Received Date: October 28, 2016; Accepted Date: November 14, 2016; Published Date: November 24, 2016
Citation: Ghafari M, Gleiter H, Feng T, Ohara K, Hahn H (2016) Are Transition Metal-Rich Metallic Glasses Made Up of Distorted Bcc Clusters?. J Material Sci Eng 5: 299. doi: 10.4172/2169-0022.1000299
Copyright: © 2016 Ghafari M, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Material Sciences & Engineering
A structural model of metal-rich amorphous alloys is proposed. This model differs from the known models of the atomic structure of amorphous metal-rich alloys consisting of transition metal-early transition metal components or of transition metal-metalloid components. The atomic structures of these kinds of alloys were studied by means of synchrotron high energy X-ray diffraction. The results obtained are presented in the form of radial distribution functions suggesting that the atomic structures of amorphous transition metal-rich alloys consist of clusters with chemically disordered and structurally distorted bcc cells, irrespective of the crystal structure of the corresponding pure metallic elements.
Amorphous alloys; Atomic structure; Radial distribution function; Transition metals; X-ray diffraction
Numerous experimental studies  of metallic glasses indicate that the arrangement of atoms in metallic glasses is aperiodic and isotropic. The presence of Short Range Order (SRO) and Medium Range Order (MRO) indicates that the atomic positions are correlated up to few interatomic distances.
Since the discovery of amorphous metallic alloys the following prominent structural models have been proposed: (a) The Dense Random Packing of Hard Spheres (DRPHS) [2-4], (b) the model based on the formation of structural units, such as trigonal prisms or icosahedral arrangements [5-9], and (c) the Cluster Model (CM) of glasses [10,11].
In fact, the analysis of the structure provides the suggestion that a uniform structure model of amorphous alloy is ineligible. In some cases, the chemical bonding is the decisive factor [12,13], then other factors such as metallic binding .
Recent reports on the structure of metal-rich amorphous alloys  as well as of the structure of isolated nanometer-sized Fe crystals  suggest that new insight into the structure of amorphous metallic materials consisting of transition metal-early transition metal components or of transition metal-metalloid components may be obtained by investigating (e.g., by means of the Radial Distribution Function, RDF of these materials as well as of some of their physical properties) the atomic arrangements in these materials. In fact, metalrich amorphous alloys consisting of transition metal-early transition metal components or of transition metal–metalloid components seem promising model systems for such studies. This view is supported by the following observations reported in the literature.
The results of Mössbauer measurements show that the distribution of the internal magnetic hyperfine fields and the quadrupole splitting distribution of pure nanometer-sized Fe crystals  and amorphous Fe90Sc10  or Fe90Zr10  are similar if the Fe crystals have a size of 2 nm or less. These observations indicate that the atomic environments of 57Fe-atoms in both materials are similar because the quadrupole splitting, which is caused by the interaction of electrical field gradient and the quadrupole moment of nucleus is directly related to the atomic environment. Furthermore, Mössbauer experiments measured in external magnetic fields and at various temperatures have verified the existence of magnetic clusters in a-Fe90Sc10 with average cluster sizes close to the interatomic spacing as deduced from the Radial Distribution Function, RDF [17,18].
Moreover, additional structural information may be obtained by studying the photon energy because the photon energy is directly related to the vibrational dynamics of the atoms and hence to their structural arrangement. The shape and the position of the maximum of the phonon DOS amorphous Fe-rich alloys  and amorphous Fe  were found to be similar to nano-crystalline-Fe  indicating that the atomic environments of 57Fe-atoms in both materials are similar
Using genetic algorithm calculations, Nguyen et al.,  have shown that by starting of random configurations, the structure of minimum-energy of the FeCo alloy is the bcc atomic arrangement. All of these observations suggest a structural similarity between crystalline Fe and amorphous Fe-rich alloys. Based on this observation, it is the aim of the study reported in this paper to investigate of the following two questions:
1- Is the atomic arrangement in transition metal–rich amorphous alloys similar to the atomic arrangement in transition metal– rich crystalline metals?
2- Does the atomic arrangement in transition metal-rich amorphous alloys depend on their chemical composition?
The materials systems chosen for the investigations that will be reported in this paper were the following two families of alloys. (1) Transition metals-early transition alloys and (2) Transition metal-rich and metalloid alloys (produced by Hitachi Metals Ltd., Vacuumschmelze GmbH & Co. KG and in our laboratory by electrochemical deposition methods). Early transitions metals (such as Sc) and metalloids (such as P, B, and Si) are known as glass forming element.
All the metallic glasses studied by us - except of amorphous Ni86P14 and Co80P20 alloys - were prepared by melt spinning. The preparation of Ni90Sc10, Co90Sc10 and Fe90Sc10 alloys was performed in a high vacuum chamber. The amorphous Ni86P14 and Co80P20 alloys were prepared by the electrochemical deposition methods [23,24].
The atomic structures of the metallic glass were studied using highenergy XRD (HE-XRD) with a photon wavelength of 0.21 Å at the beamline BL04B2 at SPring-8. The beamline BL04B2 is designed for diffraction and scattering experiments at photon energies larger than 37 keV. The main advantage of the beamline is the ability to structurally characterize disordered samples by employing a wide range of scattering vectors Q and a low instrumental background. Further advantages are small correction terms (especially for absorption correction) and reduction of truncation errors. The analysis of High-Energy X-Ray Diffraction (HE-XRD) data and the transformation from k space to the real space is described elsewhere .
Figure 1 shows the radial distribution function of crystalline bcc-Fe, fcc-Ni and hcp-Co and of the amorphous transition metal-rich Fe90Sc10 alloy. As may be seen with the similarity exists between pattern of the peaks of the RDF of amorphous Fe90Sc10 alloy and crystalline Fe with a bcc structure. However, the peaks of the amorphous alloy (Figure 1) are broader than the ones of the Fe bcc crystals and the positions of the peaks are shifted which indicates different interatomic spacing in the amorphous Fe90Sc10 alloy and crystalline Fe. The similarity between pattern of the peaks of the RDF of amorphous Fe90Sc10 alloy and crystalline Fe with a bcc structure indicates that the atomic structure of the amorphous alloy can be described as arrangements of distorted cluster with a bcc structure (the peaks of the amorphous alloy were broader than the ones of the Fe bcc crystals) and a distribution of interatomic spacing centered on those of bcc-Fe with a certain shift of the peak positions. In pure bcc-Fe, the nearest neighbor (nn) and the next nearest neighbor (nnn) atoms are located at 2.48 Å and 2.87 Å, respectively, with 8 atoms and 6 atoms as nn and nnn in the corresponding distances. Due to the small difference between the nn and nnn shells in bcc-Fe, the nn and nnn shells overlap in the amorphous alloy so that the nn and nnn shells become - in amorphous metal-rich alloy - indistinguishable.
As may be seen from Figures 2 and 3, the RDF’s of crystalline bcc-Fe and the RDF’s of amorphous Co90Sc10 and Ni90Sc10 alloys yields the same result as for Fe90Sc10 alloy. The following two main features of the peak pattern of all three alloys are as follows: (1) All the peak patterns of the RDFs are identical; (2) The major differences between them are the enhanced peak widths and the shifts of peaks of the individual amorphous metal-rich alloys relative to the peaks of the bcc-Fe. In other words, the SRO and MRO of metal-rich alloys can be described to consist of chemically disordered and structurally distorted clusters with a bcc atomic arrangement. Figure 4 shows a schematic drawing of two-dimensional atomic arrangements of the nn and the nnn in a bcc cluster and the corresponding arrangements in a distorted bcc cluster with a Sc-atom in the center. Because of the different electronegativity of Fe and Sc, the Sc-atoms have the tendency to attract the Fe- atoms.
On the other hand, it may be noted that according to the results reported above, the bcc clusters making up the amorphous structures seem not to be correlated with the crystal structure of main transition metal, i.e., bcc for Fe, fcc for Ni and hcp for Co.
In summary, the similarities of the RDFs of amorphous transition metal-rich alloys of Fe, Ni and Co, indicate that the structure of all of these alloys consist of distorted bcc clusters. For amorphous alloys with transition metal compositions larger ≥ 78 at percentage, the structure seems independent of the second components of these alloys with such as Sc, B, Si or P as presented in Figure 5. Finally, let’s address the question: Why are the distorted bcc clusters the preferred structural units of the amorphous metal–rich alloys. In order to discuss this question, the structure factor, S(Q), of amorphous Fe90Sc10 alloy as a function of Q=4Π⁄λ was compared, Figure 6  by means of the pair correlation functions with the S(Q) of molten Fe at T=1870 K and at 1830 K (the melting temperature of Fe is Tm=1811 K) and with the S(Q) of an undercooled melt of Fe at T=1750 K, 1730 K and 1670 K. As may be seen, the position of peaks in the S(Q) plots (up to about 1 Å) and the number of the nn are similar in the amorphous Fe90Sc10, in the melt and in the undercooled Fe melt. This experimental observation suggests that in rapidly quenched melts bcc clusters are the preferred structural units.
This work is supported by the National Natural Science Foundation of China (Grant no. 51571119, 51520105001), the Natural Science Foundation of Jiangsu Province and the Fundamental Research Funds for the Central Universities. T. F. acknowledges the support by the innovation project and the specially-appointed professor project of Jiangsu province. The HE-XRD Experiments have been carried out at SPring-8 at the beamline BL04B2.
Make the best use of Scientific Research and information from our 700 + peer reviewed, Open Access Journals