Scholarly article on topic 'Bulk metallic dual phase glasses by severe plastic deformation'

Bulk metallic dual phase glasses by severe plastic deformation Academic research paper on "Materials engineering"

Share paper
Academic journal
OECD Field of science
{"A metallic glass" / "B mechanical properties" / "B phase stability" / "B bonding" / "C severe plastic deformation" / "D microstructure"}

Abstract of research paper on Materials engineering, author of scientific article — Lisa Krämer, Yannick Champion, Karoline S. Kormout, Verena Maier-Kiener, Reinhard Pippan

Abstract Two different metallic glass powders were consolidated and deformed via high-pressure torsion to synthesize amorphous dual phase composites. The influence of volume fraction of the two amorphous phases and the applied shear strain was investigated. By varying the applied strain, the dimensions of the phases could be systematically varied from the micro- to the nanometer regime and at the highest applied strain even a transition to a single phase state could be observed. The study illustrates the potential of producing novel bulk metallic glasses by deformation-induced mixing which are not accessible by the classical casting route.

Academic research paper on topic "Bulk metallic dual phase glasses by severe plastic deformation"

Contents lists available at ScienceDirect


journal homepage:

Bulk metallic dual phase glasses by severe plastic deformation Ü)

Check for

Lisa Krämer3'*, Yannick Championb, Karoline S. Kormout3, Verena Maier-Kienerc, Reinhard Pippana

a Erich Schmid Institute of Materials Science, Austrian Academy of Sciences, Leoben, Austria b Univ. Grenoble Alpes, CNRS, Grenoble INP, SIMaP, F-38000, Grenoble, France c Department of Physical Metallurgy and Materials Testing, Montanuniversität Leoben, Austria



Keywords: A metallic glass B mechanical properties B phase stability B bonding

C severe plastic deformation D microstructure

Two different metallic glass powders were consolidated and deformed via high-pressure torsion to synthesize amorphous dual phase composites. The influence of volume fraction of the two amorphous phases and the applied shear strain was investigated. By varying the applied strain, the dimensions of the phases could be systematically varied from the micro- to the nanometer regime and at the highest applied strain even a transition to a single phase state could be observed. The study illustrates the potential of producing novel bulk metallic glasses by deformation-induced mixing which are not accessible by the classical casting route.

1. Introduction

Bulk metallic glasses (BMGs) engage the science communities since their discovery in 1960 [1]. Besides studies on single phase BMGs, efforts are made to control the properties by producing bulk metallic glass composites (BMGCs) [2,3]. Different combinations are possible such as amorphous/crystalline, amorphous/quasicrystalline and amorphous/ amorphous [2]. The shape of the second phase can also vary from particles over dendrites to wires and even more complex forms [4-9]. One way to produce amorphous/amorphous composites is by exploiting phase separations, which can occur during annealing in the supercooled liquid region or of the undercooled melt. This chemical demixing is caused by a positive enthalpy of mixing or as pre-stage to primary crystallization. Some well studied compositions for BMGs show this behavior as Vit105, but also Cu-, Ni-, Pd-, Zr- and Mg- based BMGs [2]. Another technique is to use two targets with different chemical composition for inert gas condensation and to produce bulk samples by in-situ compaction [10]. High-pressure torsion (HPT) was used in the past to produce and deform various materials and composites. The advantages of this technique are the high flexibility as the starting material can be either in bulk or powder form and the large variation in applied strain by simply changing the number of rotations. Due to the high, nearly hydrostatic pressure, also brittle materials [11,12] are deformable and the applied strain can go up to shear strains as high as Y = 20000 and more. This gives the possibility to produce super-

saturated solid solutions [13,14] and amorphization of crystalline phases. BMGs and BMGCs can be produced by HPT [15-26] and furthermore, it can be used for adjusting the microstructure of the composites [27] or rejuvenation of BMGs [28-33].

The aim of this study was to produce amorphous/amorphous BMGCs via HPT and to investigate the evolving microstructures. Main questions addressed are: How does the applied shear strain affect the microstructure? How do the mechanical properties change with refining of the phases? What happens if the applied shear strain goes to very high values? For this purpose, two different compositions (Ni-MG 30 wt% Zr-MG and Ni-MG 50 wt% Zr-MG) were produced as well as single phase Ni-MG and Zr-MG samples as reference. The influences of shear strain and ratio of the two phases were investigated with scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray diffraction (XRD), nanoindentation, and hardness measurements.

2. Materials and methods

The metallic glass powders (Zr57Cu20Al10Ni8Ti5 and Ni53Nb20Ti10Zn8Co6Cu3, spherical particles with diameters between 1 |jm and 40 |jm) were fabricated via high pressure gas atomization [3] and were blended in the respective composition by hand. The mixtures were then filled into the gap between two grooved steel anvils and compacted by HPT (4 GPa and 10° rotation). The main

* Corresponding author. E-mail address: (L. Krämer).

Received 10 October 2017; Received in revised form 24 November 2017; Accepted 3 December 2017

0966-9795/ © 2017 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (

deformation was applied at higher pressures (8-9 GPa) at room temperature and the rotations were varied from 2 to 100. The thickness varied from 0.45 to 0.6 mm due to the anvil preparation, but the diameter of the specimens were constant with 6 mm. The applied shear strain y can be estimated with

Y = 2n r N/t (1)

where r is the radius of the disk, N is the number of rotations and t is the thickness after deformation. SEM and hardness measurements were performed on the cross sections of the HPT disks, which were produced by wire saw cutting and further grinding and polishing. For TEM samples, small pieces at larger radius (about 2.5 mm) were cut from the disks and the samples were prepared via a standard procedure with dimple grinding and subsequent ion milling (to avoid heating, samples were cooled with liquid nitrogen) in radial direction. Structural analysis were conducted in Carl Zeiss Leo 1525 field emission scanning electron microscope using the back-scatter and in-lens detector and in a Cs-corrected JEOL JEM-2100F operated at 200 kV transmission electron microscope.

Vickers hardness was measured along the diameter on the cross section with a load of 0.5 kg and 1 kg. An error of 2% for the measured hardness values is assumed and the standard deviation errors of the fitted curve were calculated and used for the error bars in the inlet of Fig. 3.

For the nanoindentation property map, a platform nanoindenter G200 (Keysight Tec) was used and the experiments were conducted under constant indentation strain rate (0.05 s-1) to a maximal indentation depth of 100 nm with a diamond Berkovich tip (Synton-MDP Inc.). Hardness and Young's modulus were measured over indentation depth with a continuous stiffness measurement unit. All data were evaluated according to Oliver and Pharr [34] using a Poisson ration of 0.37 for the modulus and were averaged between indentation depths from 80 to 95 nm. To estimate the phase fraction for each indent, SEM micrographs with very high resolution were taken and the free software ImageJ was used to mark the Ni-MG lamellae by hand (due to the edge effect of the indents an automatic approach was not feasible) and only the regions inside the indents were evaluated. The lamellae thickness was determined by evaluating the intersections with a circle (also in the region of the respective indent), which was performed with an ImageJ plugin called "Oval Profile Plot" by Bill O'Connell [35]. Determining phase fraction and thickness with this methods leads to several errors. One is the lack of information, as the volume beneath the indent is unknown. Another is inaccurate marking of the Ni-MG phase by the author as subjective perception will distort the results (repeating the phase analysis for one indent showed a difference of 4 vol%).

A 5-circle X-ray diffractometer equipped with a source for Cu-Ka radiation was used for reflection XRD phase analysis of the Zr-MG and the Ni-MG 50 wt% Zr-MG with lower applied deformation. Only pieces from the outer part of samples were used to exclude the nearly un-deformed material in the center of the disks. Additionally, the surface was ground to remove any impurities from the HPT process. The other samples were investigated with Synchrotron X-ray diffraction performed at the PETRA III P07 beamline at the DESY Photon Science facility (Hamburg) using a beam energy of 111 keV. The primary and secondary slit of 1 x 0.25 mm2 and 1.2 x 0.3 mm2, respectively, were used to measure in a small strain range. The measured transmission diffraction patterns were analyzed with FIT 2D software. The theta angles were converted to scattering vector q with

q = 4n sin(0)/l (2)

where 0 is the reflection angle and X is the wavelength of the rays. This conversion allows plotting data from the synchrotron and lab dif-fractometer measurement in one figure.

3. Results and discussion

3.1. Microstructural evolution as function of strain

The evolution of the microstructure was investigated by SEM, TEM, and XRD. In Fig. 1a and b, SEM micrographs of two different composites (Ni-MG 30 wt% Zr-MG and Ni-MG 50 wt% Zr-MG) are shown for different applied strains. The micrographs in the second row are taken from the same position as the micrograph above but at higher magnifications and the shear direction is indicated with an arrow on the right side. As the starting materials were powders, a deformation and consolidation of the powder particles takes place in the beginning. Due to the lower hardness, this consolidation is carried mainly by the Zr-MG (light grey), while the Ni-MG particles (darker grey) show hardly any deformation and the material is hold together by the deformed softer phase (see first micrographs on the left side). The welding of the particles is not completed and the boundaries of the initial powder particles are still visible. As the two powders had similar sizes and forms, no problems regarding agglomeration was expected and at low deformations, good mixing of the two materials was observed. Regions with higher contents of Ni-MG result from the higher volume fraction of Ni-MG particles in Ni-MG 30 wt% Zr-MG, but subsequent deformation ensures a homogeneous distribution of both phases. With higher deformation, both phases are forced to deform and a lamellar structure evolves. This structure refines with increased strain and the lamella length decreases from tens of micrometers to 1 |jm and the width from 10 |jm below 10 nm. TEM micrographs in Fig. 1c, show the fine microstructure of Ni-MG 30 wt% Zr-MG at y = 1200. Thin and long lamellae are sometimes disrupted by shear bands, which cause steps and further refine the phases. At very high shear strains, a saturation is reached, where only one phase is detectable in the micrographs. A single phase microstructure imaged by SEM can be explained by a refinement of the lamellae below the resolution limit or mixing of the two MGs into a new phase. Changing the ratio of the two MGs, leads to a slightly different microstructure evolution. Ni-MG 50 wt% Zr-MG shows a fast consolidation at the beginning as the fraction of the softer phase is relatively high and it can act as glue to the harder Ni-MG particles. Therefore, no cracks are detectable even at low strains as y = 60 (see Fig. 1b). On the other hand, the high fraction of the softer Zr-MG can carry most of the applied strain and the Ni-MG is not forced to deform. This leads to an inhomogeneous deformation behavior, where large Ni-MG lamellae can still be found even at very high strains (y = 2700). If the fraction of the harder phase is higher, strains higher than y = 590 must be applied to achieve fully dense samples without cracks along the phase boundaries (see Fig. 1a). The delay of the consolidation is caused by the lower fraction of the glue-like phase, which cannot hold all Ni-MG particles together, and by the hardness of the Ni-MG, which impedes fast welding of the Ni-MG particles. The lower fraction of the softer phase forces the Ni-MG to carry more of the deformation and this causes a more homogenous deformation. The lamellae thickness is more evenly distributed and the saturation is reached at lower strains compared to Ni-MG 50 wt% Zr-MG.

In Fig. 2a, XRD profiles of the two single phase BMGs and the two composites are shown with intensity as a function of the scattering vector q. For Ni-MG + 50 wt% Zr-MG, profiles of two different samples are shown. For one profile, an intermediate microstructure was chosen. In this state, two overlapping amorphous peaks from the two initial MGs can be seen. The second profile is acquired at higher applied shear strains; only one amorphous peak is detectable and its position lays between the positions of the amorphous peaks of the initial two metallic glasses. This can be explained by generating a new amorphous phase by mixing of the two initial MGs. Generating a single phased MG consists with SEM-micrographs, where also only one phase is detectable at high applied shear strains (see Fig. 1). For Ni-MG + 30 wt% Zr-MG, only the profile at saturation is shown, where a single amorphous peak is

Fig. 1. Illustration of the microstructure of (a) Ni-MG 30 wt% Zr-MG (SEM micrographs), (b) Ni-MG 50 wt% Zr-MG (SEM micrographs) and (c) Ni-MG 30 wt% Zr-MG at y = 1200 (TEM micrographs). In (a) and (b) the top and bottom row of micrographs are taken at different magnifications. The evolution of the microstructure can be seen in dependence of the applied strain. The deformation increases from the left to the right and the microstructure changes from Ni-MG particles hold together by the softer Zr-MG to a lamellar structure. These lamellae refine with increasing applied strain until a complete mixing of the two phases occurs. Take notice in the increase of applied strain for higher contents of Zr-MG to reach saturation. The evolving fine lamellar structure can be seen in (c) and lamellae thickness decreases below 10 nm.

Fig. 2. (a) XRD results of different composite concentrations. For Ni-MG 50 wt% Zr-MG, samples with transient microstructure was measured and two overlapping peak of the initial MGs can be seen as well as a rising peak in between from the new MG. At saturation, both composites show just one remaining amorphous peak. (b) Positions of the first amorphous peak from XRD measurements depending on the composition. The position correlates linearly with the chemical composition of the composites. Higher fractions of Ni shifts the peak to higher q values.

detectable. The saturation peak position depends on the composition and shifts linearly with the percentage of the phases. This can be seen in Fig. 2b, where the peak position is plotted depending on the vol% of Ni-MG. Higher vol% of Ni-MG means a higher fraction of Ni in the metallic glass and the position shifts linearly to higher q values.

3.2. Influences of the microstructure on mechanical properties

The microhardness as a function of the strain is shown in Fig. 3. The plotted lines should help to guide the eye and have no physical meaning. Zr-MG shows a nearly constant hardness as a function of the applied strain and the large scattering at small strain can be explained by an uncomplete welding of the particles, where cracks and inhomogeneities reduce the hardness. Ni-MG shows a slight softening with higher applied shear strain.

Zr-MG and Ni-MG have a saturation hardness of 470 and 640 HV, respectively. In comparison, the hardness of the composites increases with the applied strains at the beginning, but then flattens until a saturation is reached. For Ni-MG 30 wt% Zr-MG, the scatter of the hardness at low strains is very high, which is caused by the inhomogeneous microstructure with cracks and undeformed particles (see micrographs in Fig. 1a and b at Y = 180), compared to Ni-MG 50 wt% Zr-MG. In the inlet, the approximated saturation values are plotted over the vol% of Ni-MG. The saturation hardness of the two composites lays between the hardness of the two initial MGs and depends directly on the fraction of the two MGs, but is higher than the rule of mixture predicts. In literature two different explanation can be found for an increased hardness level induced by compositional changes. Deng et al. supposed the presence of a harder crystalline phase for some compositions [36]. However, crystallization even on the nanoscale was

Fig. 3. Illustration of the Vickers microhardness evolution as a function of the shear strain. The hardness of the single phase BMGs is constant over the applied strain, whereas the composites show a strong increase in hardness at low strains and then a saturation. The saturation hardness lays higher than the rule of mixture predicts (black line in the inlet).

Fig. 4. (a) Illustration of a nanoindentation property map, graphs of (b) the Young's modulus (blue) and hardness (green) depending on the phase fractions in the impressions, and (c) the Young's modulus and hardness depending on the lamellae thickness. The residual impressions in (a) are colored respective to their measured hardness and for four indentation marks, micrographs with colored phases are included. Graphs in (b) and (c) show that the hardness and Young's modulus depend on the volume fraction of the phases, but not on the lamellae thickness.

excluded for this study by TEM investigations. Donald et al. showed a correlation of the hardness with the average outer electron concentration by adding different elements [37]. The composition changes drastically by mixing of the two amorphous phases and for this reason, a change in the outer electron concentration can be assumed. This can explain the discrepancy between the actual hardness and the one the rule of mixture predicts. The higher hardness of the newly formed amorphous phase causes the increase in hardness for intermediate strain regime, but a size effect cannot be ruled out. Thinning of the lamellae and increasing the number of interfaces can also influence the hardness and further studies must be conducted.

Nanoindentation was used to investigate the influence of the lamella

thickness and the ratio of the two phases. In Fig. 4a, a micrograph with a properties map with three times eight indentations is shown. The residual imprints are colored according to the measured nanohardness and the scale bar can be seen on the bottom. Additionally, colored micrographs for phase analysis are added for some indentation marks. At first glance, it can be seen that the large Ni-MG particle has a higher hardness than the surrounding area, in which Zr-MG and Ni-MG contribute to the observed hardness. In Fig. 4b, the hardness and the Young's modulus of the indentations are plotted as a function of the measured volume fraction of the Ni-MG, which was calculated from SEM micrographs as shown in the small inlets in Fig. 4a. Additionally, average values from reference nanoindentation data of the two single

phase and two composites at saturation are plotted for comparison reasons. The trend shows that for both, properties map and reference indentations, Young's modulus and hardness increase with increasing volume fraction of the Ni-MG, which is harder and stiffer than the Zr-MG. One challenge of measuring the volume fraction is the mixing of the two phases, which already occurs in the finer structured regions of the map. This directly leads to an underestimation of the volume fraction of Ni-MG in this region and also to a corresponding increase of the hardness as the new phase has a higher hardness, which can be also seen in comparison with reference indentations in the composite structures at saturation. The lamella thickness of the Ni-MG does not have a significant influence on the hardness, which can be seen by comparing the fifth indentation of the first row and the first indentation of the third row. Both show nearly the same hardness (10.23 and 10.29 GPa), although the one in the first row indents mainly a large lamella whereas the second one probes a region with a very fine microstructure. In Fig. 4c, the nanohardness and Young's modulus are plotted versus the thickness of the Ni-MG lamellae and no significant correlation is obvious. Therefore, no size effect larger than the systematic measurement error of the lamellae thickness was detected. This can be explained by TEM micrographs (Fig. 1c), where shear bands propagate through both phases without deflection at the boundaries. For this system, phase boundaries provide no barrier to shear bands and therefore no increase in hardness is observed at smaller lamellae. Other systems that comprise amorphous phases with different properties (different Poisson ratios) might show a size effect.

4. Conclusion

In summary, it was successfully shown that HPT can be used to generate amorphous/amorphous composites in bulk size and the microstructure can be controlled by the applied strain. The lamellar structure is tunable from the micrometer to the nanometer regime and at very high strains, even mixing of the two amorphous materials towards single phased BMGs occurs. The ratio of the two initial amorphous phases is smoothly adaptable, because powders were used as starting materials and therefore BMGs with different chemical compositions are producible. Higher fractions of the softer phases enhances consolidation and a crack free microstructure is reached at low strains. It also hinders a homogenous deformation and the saturation shifted to higher strains. Hardness measurements show that mixing and ratio of the two phases have a stronger influence on the mechanical properties than the microstructure. Hardness and Young's modulus correlates with the ratio of the two phases, but no clear size effect was detectable. At saturation, the newly formed amorphous phases have a hardness between the initial MGs, but higher than the rule of mixture predicts.


The European Research Council under ERC Grant Agreement No. 3401 85 USMS has provided funding for this work. The financial support by the Austrian Federal Government (837900) within the framework of the COMET Funding Program (MPPE, A7.19) is highly appreciated.


[1] W. Klement, R.H. Willens, P. Duwez, Non-crystalline structure in solidified goldsilicon alloys, Nature 187 (1960) 869-870,

[2] J. Eckert, J. Das, S. Pauly, C. Duhamel, Mechanical properties of bulk metallic glasses and composites, J. Mater. Res. 22 (2007) 285-301, 1557/jmr.2007.0050.

[3] L. Perrière, Y. Champion, Phases distribution dependent strength in metallic glass -aluminium composites prepared by spark plasma sintering, Mater. Sci. Eng. A 548 (2012) 112-117,

[4] H. Choi-Yim, R. Busch, U. Köster, W. Johnson, Synthesis and characterization of particulate reinforced Zr57Nb5Al10Cu15.4Ni12.6 bulk metallic glass composites, Acta Mater. 47 (1999) 2455-2462,


[5] R.D. Conner, R.B. Dandliker, W.L. Johnson, Mechanical properties of tungsten and steel fiber reinforced Zr41.25Ti13.75Cu12.5Ni10Be22.5 metallic glass matrix composites, Acta Mater. 46 (1998) 6089-6102,

[6] S.T. Deng, H. Diao, Y.L. Chen, C. Yan, H.F. Zhang, A.M. Wang, Z.Q. Hu, Metallic glass fiber-reinforced Zr-based bulk metallic glass, Scr. Mater. 64 (2011) 85-88,

[7] Y.-K. Xu, H. Ma, J. Xu, E. Ma, Mg-based bulk metallic glass composites with plasticity and gigapascal strength, Acta Mater. 53 (2005) 1857-1866, http://dx.doi. org/10.1016/j.actamat.2004.12.036.

[8] T. Liu, P. Shen, F. Qiu, Z. Yin, Q. Lin, Q. Jiang, T. Zhang, Synthesis and mechanical properties of TiC-reinforced Cu-based bulk metallic glass composites, Scr. Mater. 60 (2009) 84-87,

[9] H. Shakur Shahabi, S. Scudino, U. Kühn, J. Eckert, Metallic glass-steel composite with improved compressive plasticity, Mater. Des. 59 (2014) 241-245, http://dx.

[10] N. Chen, D. Wang, T. Feng, R. Kruk, K. Yao, D.V. Louzguine-Luzgin, H. Hahn,

H. Gleiter, A nanoglass alloying immiscible Fe and Cu at the nanoscale, Nanoscale 7 (2015) 6607-6611,

[11] K. Edalati, M. Arimura, Y. Ikoma, T. Daio, M. Miyata, D.J. Smith, Z. Horita, Plastic deformation of BaTiO3 ceramics by high-pressure torsion and changes in phase transformations, optical and dielectric properties, Mater. Res. Lett. 3 (2015) 216-221,

[12] H. Razavi-Khosroshahi, K. Edalati, M. Arita, Z. Horita, M. Fuji, Plastic strain and grain size effect on high-pressure phase transformations in nanostructured TiO 2 ceramics, Scr. Mater. 124 (2016) 59-62, 2016.06.022.

[13] K.S. Kormout, R. Pippan, A. Bachmaier, Deformation-induced supersaturation in immiscible material systems during high-pressure torsion, Adv. Eng. Mater. 19 (2017) 1600675, ,

[14] A. Bachmaier, J. Keckes, K.S. Kormout, R. Pippan, Supersaturation in Ag-Ni alloy by two-step high-pressure torsion processing, Philos. Mag. Lett. 94 (2014) 9-17,

[15] A.R. Yavari, W.J.B. Filho, C.A.D. Rodrigues, C. Cardoso, R.Z. Valiev, Nanostructured bulk Al90Fe5Nd5 prepared by cold consolidation of gas atomised powder using severe plastic deformation, Scr. Mater 46 (2002) 711-716, http://dx.

[16] J. Sort, D.C. Ile, A.P. Zhilyaev, A. Concustell, T. Czeppe, M. Stoica, S. Surinach,

J. Eckert, M.D. Baro, Cold-consolidation of ball-milled Fe-based amorphous ribbons by high pressure torsion, Scr. Mater 50 (2004) 1221-1225, 1016/j.scriptamat.2004.02.004.

[17] A. Révész, S. Hobor, P.J. Szabo, A.P. Zhilyaev, Z. Kovacs, Deformation induced crystallization in an amorphous Cu60Zr20Ti20 alloy by high pressure torsion, Mater. Sci. Eng. A 460-461 (2007) 459-463, 2007.01.081.

[18] J.B. Fogagnolo, M.F. De Oliveira, C.S. Kiminami, C. Bolfarini, W.J.F. Botta, Consolidation of easy glass former Zr55Cu30Al 10Ni5 alloy ribbons by severe plastic deformation, J. Metastab. Nanocryst. Mater. 20-21 (2004) 253-256, http://

[19] A.A. Kündig, T. Schweizer, E. Schafler, J.F. Löffler, Metallic glass/polymer composites by co-processing at similar viscosities, Scr. Mater 56 (2007) 289-292,

[20] Y.F. Sun, H. Fujii, N. Tsuji, Y. Todaka, M. Umemoto, Fabrication of ZrAlNiCu bulk metallic glass composites containing pure copper particles by high-pressure torsion, J. Alloys Compd. 492 (2010) 149-152, 11.135.

[21] Y.F. Sun, T. Nakamura, Y. Todaka, M. Umemoto, N. Tsuji, Fabrication of CuZr(Al) bulk metallic glasses by high pressure torsion, Intermetallics 17 (2009) 256-261,

[22] K. Edalati, Y. Yokoyama, Z. Horita, High-pressure torsion of machining chips and bulk discs of amorphous Zr50Cu30Al10Ni10, Mater. Trans. 51 (2010) 23-26,

[23] X. Sauvage, Y. Champion, R. Pippan, F. Cuvilly, L. Perrière, A. Akhatova, O. Renk, Structure and properties of a nanoscaled composition modulated metallic glass, J. Mater. Sci. 49 (2014) 5640-5645,

[24] H. Asgharzadeh, S.H. Joo, J.K. Lee, H.S. Kim, Consolidation of Cu-based amorphous alloy powders by high-pressure torsion, J. Mater. Sci. 50 (2015) 3164-3174, http://

[25] J. Qiang, K. Tsuchiya, Concurrent solid-state amorphization and structural rejuvenation in Zr-Cu-Al alloy by high-pressure torsion, Mater. Lett. 204 (2017) 138-140,

[26] L. Krämer, K. Kormout, D. Setman, Y. Champion, R. Pippan, Production of bulk metallic glasses by severe plastic deformation, Met. (Basel) 5 (2015) 720, http://dx.

[27] L. Krämer, Y. Champion, R. Pippan, From powders to bulk metallic glass composites, Sci. Rep. 7 (2017) 6651,

[28] Y.B. Wang, D.D. Qu, X.H. Wang, Y. Cao, X.Z. Liao, M. Kawasaki, S.P. Ringer, Z.W. Shan, T.G. Langdon, J. Shen, Introducing a strain-hardening capability to improve the ductility of bulk metallic glasses via severe plastic deformation, Acta Mater 60 (2012) 253-260,

[29] X.D. Wang, Q.P. Cao, J.Z. Jiang, H. Franz, J. Schroers, R.Z. Valiev, Y. Ivanisenko, H. Gleiter, H.J. Fecht, Atomic-level structural modifications induced by severe plastic shear deformation in bulk metallic glasses, Scr. Mater 64 (2011) 81-84,

[30] F. Meng, K. Tsuchiya, I. Seiichiro, Y. Yokoyama, Reversible transition of

deformation mode by structural rejuvenation and relaxation in bulk metallic glass, Appl. Phys. Lett. 101 (2012) 121914, ,

[31] S. Joo, D. Pi, A.D.H. Setyawan, H. Kato, M. Janecek, Y.C. Kim, S. Lee, H.S. Kim, Work-Hardening induced tensile ductility of bulk metallic glasses via high-pressure torsion, Sci. Rep. 5 (2015) 9660,

[32] J. Das, M.B. Tang, K.B. Kim, R. Theissmann, F. Baier, W.H. Wang, J. Eckert, "work-hardenable" ductile bulk metallic glass, Phys. Rev. Lett. 94 (2005) 205501,, http://

[33] J. Bünz, T. Brink, K. Tsuchiya, F. Meng, G. Wilde, K. Albe, Low temperature heat capacity of a severely deformed metallic glass, Phys. Rev. Lett. 112 (2014) 135501, ,

[34] W.C. Oliver, G.M. Pharr, An improved technique for determining hardness and

elastic modulus using load and displacement sensing indentation experiments, J. Mater. Res. 7 (1992) 1564-1583,

[35] B. O'Connell, Oval Profile Plot, (n.d.). (Accessed August 18, 2017).

[36] Y.P. Deng, Y.F. Guan, J.D. Fowlkes, S.Q. Wen, F.X. Liu, G.M. Pharr, P.K. Liaw, C.T. Liu, P.D. Rack, A combinatorial thin film sputtering approach for synthesizing and characterizing ternary ZrCuAl metallic glasses, Intermetallics 15 (2007) 1208-1216,

[37] I.W. Donald, H.A. Davies, The influence of transition metal substitutions on the formation, stability and hardness of some Fe- and Ni-based metallic glasses, Philos. Mag. A Phys. Condens. Matter, Struct. Defects Mech. Prop. 42 (1980) 277-293,