Viscosity is one of the most important parameters to characterize a liquid. In the supercooled state as temperature is lowered it increases rapidly, by as much as 15 orders of magnitude or more, before the sample reaches the glassy state. The microscopic origin of viscosity and the reason why it increases so much in the supercooled region have long been a subject of intense debate because it is a key question regarding the formation of a glass. Recently we found that the origin of viscosity at high temperatures is the local configurational fluctuations. We have shown that the Maxwell relaxation time, tau;_M (= eta;/G, where eta; is viscosity and G is the high-frequency shear modulus), is equal to the time-scale of local configurational excitation (LCE), tau;_LC, defined as the time to lose or gain one nearest neighbor [1]. Thus a macroscopic quantity, tau;_M, is directly related to a microscopic quantity, tau;_LC. This relation, however, breaks down below the crossover temperature, TA, where viscosity shows a super-Arrhenius behavior. We found that this is because transverse phonons are localized above TA, so that LCEs cannot communicate each other and thus they are independent of each other. Below TA they communicate through the dynamic stress field they create, resulting in tau;_M > tau;_LC. This explains the super-Arrhenius behavior and rapid increase in viscosity as temperature approaches Tg, the glass transition temperature. These results represent a view of atomic dynamics in the liquid state very different from the conventional one. We are currently working on the experimental verification of the relation, tau;_M = tau;_LC, using inelastic neutron and x-ray scattering.
[1] T. Iwashita, D. M. Nicholson and T. Egami, Phys. Rev. Lett., 110, 205504 (2013).

The properties of metallic glasses depend on their structures. Here we demonstrate the existence of backbone structures in metallic glasses through computational studies focusing on metal-metalloid systems. The backbone of metallic MGs distinguishes itself by its topological and chemical ordering in the medium range. The atomic-level structure of the backbone can be best described with the concept of tetrahedral cluster packing, where the packing of local structural motifs is highly efficient. The backbone structure is found to be inherited from the supercooled liquid, and shows no signs of spatial growth during the glass transition. The spatial extension and dimensionality of the backbone is highly correlated with geometrical frustration limit of cluster packing. By applying metadynamics to alter the structure of the backbone, new glasses with distinct energetics and structures is found. The existence of backbones has profound effects on the properties of metallic glasses. We further shed light on (1) the heterogeneous relaxation dynamics of the supercooled liquid by correlating it with the formation of the backbone structure; and (2) the devitrification process of metallic glasses and the destabilization of the backbone.

Despite the lack of long-range order in metallic glasses (MGs) it is thought that significant localized atomic ordering exists. At the length scale of nearest-neighbor arrangements, the short-range atomic order (SRO) is dictated by efficient atomic packing in clusters or polyhedra [1]. The packing of these quasi-equivalent clusters then gives rise to a distinct medium-range atomic order (MRO). Zr_xCu_100minus;x MGs have good glass-forming abilities over many compositions [2]. Modeling studies of this system using molecular dynamics [3, 4] and reverse Monte Carlo [5] have identified the dominant SRO as Cu-centered icosahedral clusters using Voronoi analysis [3, 5] and a cluster template method [4]. The incidence of icosahedra increases with increasing Cu [3, 5]. The short-range stability of icosahedral clusters in the liquid [6] contributes greatly to the slowing structural dynamics at the glass transition [3]. Icosahedral clusters in the models demonstrate a strong spatial correlation compared to other polyhedra, suggesting a string-like icosahedral MRO [5] that might determine strength and brittleness.
In the present study we measure and map the magnitude of angular correlations in an array of scanning electron nano-diffraction (SEND) patterns obtained from a Zr₃₆Cu₆₄ glass [7]. By tuning the coherence length of the electron probe to the size of the SRO clusters in the glass, the angular symmetries in the SEND patterns reflect prominent symmetries of the SRO clusters. We statistically analyze the incidence of prominent two, six and ten-fold symmetries in the SEND patterns of this glass, and identify the predominant SRO in these materials as icosahedral clusters, consistent with many modeling studies [3,4,5]. By mapping the magnitude of these symmetries we infer that the MRO in this material consists of face-sharing or interpenetrating icosahedra in keeping with efficient space-filling. This novel technique of mapping the strength of angular correlations in SEND patterns is a promising direct method to probe the structure of disordered materials [7].
[1] D. B. Miracle, Nature Mat., 3, 697 (2004).
[2] N. Mattern, A. Schöps, U. Kühn, J. Acker, O. Khvostikova, and J. Eckert, J. Non-Cryst. Solids, 354, 1054 (2008).
[3] Y. Q. Cheng, H. W. Sheng, and E. Ma, Phys. Rev. B, 78, 014207 (2008).
[4] X. W. Fang, C. Z. Wang, Y. X. Yao, Z. J. Ding, and K. M. Ho, Phys. Rev. B, 82, 184204 (2010).
[5] M. Li, C. Z. Wang, S. G. Hao, M. J. Kramer, and K. M. Ho, Phys. Rev. B, 80, 184201 (2009).
[6] F. C. Frank, Proc. R. Soc. Lond. A, 215, 43 (1952).
[7] A. C. Y. Liu, M. J. Neish, G. Stokol, G. A. Buckley, L. A. Smillie, M. D. de Jonge, R. T. Ott, M. J. Kramer, and L. Bourgeois, Phys. Rev. Lett., 110, 205505 (2013)

Bulk amorphous steels (BAS&’s) are one of the promising advanced materials with superior mechanical and physical properties compared with their crystalline counterparts. These unique properties make BAS&’s suitable candidates for structural applications. Although there are diverse production techniques from high purity constituent elements, however no research have been published about the synthesis of BAS&’s from scrap materials via commercially available production techniques. According to this perspective, the usage of steel scraps as a base material in the production of BAS&’s had to be investigated.
In this study, the BAS&’s consisting of cast iron scrap (having 3.5 - 4.5% C) with elemental addition (Cr, Mo, B, Ln (=Lanthanides), Y, V, W) were produced by arc melting equipped with suction casting unit as well as conventional centrifugal casting machine. In this ongoing research, 2.5 mm thick amorphous samples with having wide supercooled liquid region (T_g = 545.2°C, T_x = 663.1°C, T_m = 1136.5°C, T_rg = 0.48 and ΔT_x = 117.9°C), were produced with conventional centrifugal casting machine. The effects of alloying additions on physical, mechanical and structural properties were investigated. Nanocrystallization characteristics and kinetics will also be studied by annealing the BAS&’s in the vicinity of T_g and T_x temperatures. Both amorphous and nanocrystallized samples will be characterized by X-ray diffraction (XRD), scanning electron microscopy (SEM), differential scanning calorimerty (DSC), vibrating sample magnetometer (VSM) and mechanical testing methods.

Molecular dynamics simulation can be used to test the microscopic assumptions of constitutive theory. We have applied these techniques to investigate theories of cavitation and deformation in metallic glass. The rate of cavitation was measured in systems subjected to various degrees of hydrostatic loading. We observed two regimes of cavitation, an early regime in which classical nucleation theory holds and a later regime in which shear transformation zone (STZ) activity interferes with the cavitation process. This leads to interesting questions regarding how cavitation and STZ activation compete within the fracture process zone to determine metallic glass ductility. Furthermore we have applied MD techniques to attempt to identify STZ activation in metallic glass under differently oriented shear loading. Attempts to correlate STZ activity with normal mode analysis of the structure have proven inconclusive, but the orientation dependence of the STZ activity reveals important characteristics of the STZs and the history dependence of their distribution.

Deformation in glasses proceeds differently than in crystalline materials due to the absence of defined lattice planes and due to the absence of line defects with discrete Burgers vectors. Experiments have shown that deformation exceeding the elastic range is mostly localized in plate-like mesoscopic defects, so-called shear bands. Although the occurrence of shear bands during plastic deformation of metallic glasses is well known, their actual physical properties remain fairly unknown. Here, experimental data on the rate of atomic diffusion within shear bands has been obtained using the radiotracer method on post-deformed specimens. The experimental results indicate unambiguously that the diffusivity is largely enhanced as compared to volume diffusion in the same metallic glass at identical temperatures. This result is also discussed with respect to nanocrystal formation in shear bands.
In order to analyze the local properties of glassy matter within the shear band regions, a new approach based on analytical transmission electron microscopy methods has been developed. In fact, this approach allows to quantitatively determine the local mass density, composition and structural states with nanometer resolution. Thus, shear bands and the surrounding matrix can be analyzed separately and comparatively. In addition, to analyze the shear band propagation and interaction on a mesoscopic scale, digital image correlation in combination with the application of nanoscale marker patterns and scanning electron microscopy detection have been utilized. The experimental results are discussed with respect to the underlying mechanism during the early stages of shear band activation.

Although it is well known that significant heating can occur in active shear bands, the degree of heating and the circumstances under which heating occurs remain controversial. We have used the fusible tin coating method to detect shear band heating in amorphous Zr₅₇Ti₅Cu₂₀Ni₈Al₁₀ loaded under quasi-static uniaxial compression. By lithographically patterning the tin coating into 15 µm wide lines we can measure the shear offset as a function of position along a shear band, and simultaneously assess the amount of heat release by measuring the distance away from the shear band that melting of the tin is observed. These observations are further correlated with high-rate load data acquired with a piezoelectric load cell and high-rate video, which allow a precise determination of the time scale associated with the shearing events. On samples loaded to fracture we observe evidence of melted tin consistent with significant heat release. On samples where loading was halted prior to fracture we see no evidence of melted tin, despite the presence of shear offsets up to ~6 µm on some shear bands. We discuss the implications of these observations for our understanding of thermal evolution in shear bands, including the possibility that there is less heat released than current models predict.

Mechanical properties in nano-sized metallic glasses have been widely investigated to clearly understand the fundamental deformation mechanism of metallic glasses. In particular, the change of yield strength and the occurrence of deformation mode transition (heterogeneous to homogeneous) with sample size reduction have been reported in previous experimental studies through the nanopillar test results. Research on nano-mechanical responses of metallic glasses could lead to an in-depth understanding of the deformation mechanism related to the nucleation and propagation of shear bands. However, during the pillar fabrication using focused ion beam for nanopilar test, Ga+ ion beam damage on pillar surface is difficult to avoid and the fabrication takes relatively long for multiple pillar preparations. To avoid the drawbacks of nanopillar tests, in the present study we performed nanocompression tests of metallic glass nanoparticles. In our experiments, metallic glass nanoparticles have been fabricated by the dealloying technique from phase separated metallic glasses. From the two-amorphous alloys with droplet structure, the reactive matrix phase has been selectively dissolved by chemical process. The shape of the remaining particle phases was clearly spherical, and the sizes of fabricated particles have distribution in 40~300 nm diameter range. The effect of particle size on mechanical properties of fabricated amorphous particles has been investigated from the compression test using in-situ compression tester with electron microscope imaging. Due to the distinct morphology of particles compared to that of nanopillars, the stress-strain relations from particle compression test results are derived based on contact mechanics theory and contact area calculation. The electron beam irradiation effect during in situ compression test will be also reported, with a focus on unloading process response of nanoparticles. These results provide us with insights on evaluation of nano-mechanical properties and understanding of fundamental deformation mechanism of metallic glasses.

Tracking of the individual particles by confocal microscopy during deformation of a colloidal glass makes it possible to follow the local strain tensor as a function of the macroscopically imposed strain. Autocorrelation of this strain field reveals the distinct quadrupolar character of a sheared Eshelby inclusion. The volume-strain product of these inclusions is similar to that measured experimentally in metallic glasses. Weaker quadrupolar autocorrelations are also observed in undeformed, quiescent colloidal glasses, which indicates that Eshelby shear modes are already thermally active. The shear modulus of a colloidal glass can be measured from the probability distribution of the thermally activated shear strain energy density. As expected for these dense hard-sphere glasses with a large pressure head above them, the shear modulus increases strongly with depth (from 1 to 6 Pa). Deformation causes the shear modulus throughout the sample to drop uniformly to 1-2 Pa. After the deformation stops, the modulus relaxes back to its original profile. In dense hard-sphere systems, the modulus is a sensitive probe of the density, and hence these experiments can be used to track the small density changes associated with the deformation process.

While elastic behavior of metallic glasses are well known and documented, nonlinear response remains largely untouched, especially its relation to symmetry change and yielding. Considering that the elastic limit is typically 2-3% , one must consider the nonlinear effect into consideration in mechanical properties for most metallic glasses. In this work, we show that indeed, nonlinear effect can not be ignored as shown by the large third order elastic constants. And furthermore, we show the emergence of symmetry breaking in metallic glasses caused by external loading which is exhibited in the change of the elastic constants. The nonlinear elasticity allows us also to look into the very question of the theoretical strength of metallic glasses. We show that the tensile and compressive strength can reach E/5 and E/7 respectively, where E is the Young's modulus. Other implications of the nonlinear elastic responses in the overall mechanical properties of metallic glasses will be discussed.

All bulk metallic glasses exhibit a large and almost universal elastic strain limit. Here we show that the magnitude of the yield strain of the glass state can be quantitatively derived from a characteristic property of the flow state typical in running shear bands (the root cause of yielding). The strain in the shear flow is mostly plastic, but associated with it there is an effective elastic atomic strain. The latter is almost identical for very different model systems in our molecular dynamics simulations, such that the corresponding yield strain is universal at any given homologous temperature.

Multiple primary-damage-state cascade simulations were performed on a well
relaxed 50/50 binary model glass. As a function of the number of primary
knock-on atoms (PKAs), the resulting atomic-scale structure was characterized
in terms of the first and second derivatives of the potential energy landscape.
Local quantities such as volume, stress and Kelvin elastic moduli were
investigated, as well as the vibrational properties of the evolving structure.
Structural excitations were also investigated using the Activation-Relaxation-Technique (ART),
giving barrier energy distributions as a function of PKA number. The work finds that
on average the glass's structural features change little and rapidly saturate with respect
to increasing PKA number, with the main effect being a change in the pressure (and therefore volume)
as well as shift of the ART-derived barrier energies to lower values. These
results are discussed in terms of their implication to plasticity for small-scale
bulk metallic glass samples.

In contrast to pure metals and most non-glass forming alloys, metallic glass formers are moderately strong liquids. In general, the kinetic fragility increases with the complexity of the alloy with differences between the alloy families, e.g. noble-metal based alloys being more fragile than Zr-based alloys. Experimental kinetic and thermodynamic data are assessed for several bulk metallic glasses-forming and analyzed using the Vogel-Fulcher-Tammann (VFT) equation and the Adam-Gibbs (AG) approach. The thermodynamic temperature where the configurational entropy vanishes in the AG-approach is the same temperature where viscous flow becomes infinite in the VFT-kinetic model, suggesting the same microscopic origin and rigorously connecting the kinetic and thermodynamic aspects of fragility. Moreover, the stronger the glass, the larger is the AG-free enthalpy barrier to cooperative rearrangements, advocating the concept that in strong liquids the flow events are more localized and therefore higher activation barriers need to be overcome.

One of the more surprising developments in the area of metallic glassformers in recent years has been the suggestion by Zhang et al (J.Chem. Phys. 133, 014508 (2010). that a large number of metallic glassformers exhibit some sort of structural transition, manifested as a change from fragile liquid character at high temperature to strong liquid character at lower temperature approaching the glass transition. This behavior was previously considered a characteristic only of certain highly anomalous "tetrahedral liquids" like water, silica, BeF2 and elemental silicon. That such phenomena may also be associated with metal-to-semiconductor transitions was suggested long ago by Turnbull and coworkers in relation to silicon phenomenology, and then confirmed by the simulations of Grabow, Sastry and their coworkers.
In the metallic glassformer case, of course, the transition cannot be assigned to any electron localization phenomenon, while in the water and silica cases, there is no hint of delocalized electrons, so we must search for some more general explanation. We find this in the phenomenology of the glassformer analog class of materials called the rotator phases, which also exhibit the pattern of behavior known as strong/fragile in the case of liquids. However, in the case of plastic crystals the "strong" extreme can be associated with the occurrence of order-disorder ("lambda") transitions at higher temperatures. These critical phenomena have a well-understood theoretical underpinning, and it will be suggested that they might be relevant to the metallic glassformer behavior as well. In liquids, the transition may be damped out before becoming critical, or may be prempted by a first order transition, which might signal a sudden crystallization. In the case of certain model liquids, there is a true critical point, at which two liquids of the same composition but different density, become indistinguishable. We link these ideas to known anomalies in supercooled liquid Te, and As2Te3, and wonder if they might find practical manifestation in the fast switching phenomena of "phase change" materials. (Angell, in Physics and Applications of Disordered Materials (editor M. Popescu) pp 1-18 (2002)).

Master alloys of composition Pt57.3Cu14.6Ni5.3P22.8, Pt42.5Cu27Ni9.5P21, and Pt60Cu16Co2P22 have been produced by melting the raw elements in an induction furnace. The master alloys have been consequently purified by melting them under a flux of B2O3 and used to cast bulk metallic glass specimens, of 3 mm thickness, using a tilt-casting device and Cu-molds. For each alloy composition, thermophysical properties, both thermodynamic and kinetic, have been measured using differential scanning calorimetry (DSC) and thermal mechanical analysis (TMA), respectively. The specific heat capacities of the liquid, crystalline and glassy states are measured in DSC and the excess and configurational entropy functions of the undercooled liquid are thereafter determined. Equilibrium viscosities below the glass transition are measured using a three-point beam-bending method and the kinetic fragility parameters are determined using the empirical Vogel-Fulcher-Tamman equation. Additionally, the temperature dependence of the equilibrium viscosity is analyzed with the Adam-Gibbs configurational entropy model and an activation barrier for cooperative rearrangement are determined. Particular attention has been devoted to how these properties change with the alloy group, composition and complexity.

We have used in-situ heating in the scanning transmission electron microscope (STEM) to measure the structural evolution and glass transition temperature T_g of ZrCuAlNi bulk metallic glass thin films deposited by sputtering. Annealing at 400 °C for 2 hours increases the Zr and Cu compositional fluctuations from around 3 at. % to about 10 at. % measured with sub-nanometer electron probes on 20 nm thick Zr₅₁Cu₃₂Al₉Ni₈ films. When Zr and Cu atoms are more separated, two distinct diffraction maxima appear, indicating that the short-range order becomes less homogeneous above 350 °C. The degree and type of the medium-range order structure also shift above 400 °C. All of these correlated changes in composition and structure begin well below the T_g, and grow more pronounced until the onset of crystallization above the glass transition. T_g of Zr₅₃Cu₂₉Al₁₂Ni₆ was measured in situ using the plasmon peak energy, E_p, in electron energy loss spectroscopy (EELS). E_p is sensitive to the specific volume, V, through the number density of valence electrons. V derived from E_p as a function of temperature up to 500 °C shows two linear regions. The low temperature region corresponds to a volume thermal expansion coefficient α of 5 × 10^-5 K^-1. At high temperatures, α = 4 × 10^-4 K^-1. The intersection of the two lines gives Tg = 490 °C. This value is higher than the calorimetric T_g of 470 °C, pointing to the importance of obtaining a true reference temperature for the thin TEM sample, not just the thermocouple on the TEM sample holder. Efforts to combine this technique and temporal fluctuation electron microscopy to study nanoscale structural relaxation near T_g will also be discussed.
We acknowledge support from the National Science Foundation (LH and PMV DMR-0905793 and PKL DMR-0909037, CMMI-0900271, and CMMI-1100080).

Efforts to stabilize nanocrystalline metals through alloying share both a philosophical goal and some empirical similarities with glass forming science. The basic premise, namely that the local atomic clusters in an alloyed grain boundary may be similar to those in a comparably alloyed glass, suggests that grain boundary segregation and glass formation would have similar physical roots. Hence, thermodynamic models developed to describe the energy of grain boundaries should be able to shed a new perspective on the glass forming ability of binary alloys. In this talk we explore this analogy quantitatively for both equilibrium and non-equilibrium processing scenarios, with emphasis on its relation to classical heuristics for predicting glass forming ability: atomic size difference, heat of mixing, etc.

The influence of atomic structure on glass-forming ability and thermal stability in ternary metallic glasses is assessed for a broad range of ternary metallic glasses. Over 1000 ternary BMGs from over 100 ternary systems reported in the literature are studied. The atomic structure is quantified for each alloy using the efficient cluster-packing model, and the local atom packing efficiency is quantified around all three elements in the glass. Other data taken from the literature for these glasses includes density, maximum amorphous thickness, and thermal stability parameters such as glass transition temperature, crystallization temperature, melting temperature and liquidus temperature. Correlations are made between the atomic structure, and especially local atomic packing efficiency, with amorphous thickness and thermal stability. Topological similarities and differences between the best ternary BMGs are identified, and preferences for particular combinations of radius ratios and for solute-rich vs solute-lean glasses are reported. The full range of glasses studied, along with the detailed characterization of atomic structures, give clear trends and new insights into the nature of glass-forming ability in ternary glasses.

The glass-forming ability (GFA) or critical cooling rate of metal alloys determines the maximum thickness of metallic glass samples. Understanding the key physical properties that determine the GFA is essential for developing stronger and less costly bulk metallic glasses (BMGs). However, it is still extremely difficult to predict the GFA of multi-component alloys, and BMG design is currently a trial-and-error process. One heuristic guiding principle is that the GFA is enhanced for systems with densely packed amorphous structures. However, we find that the jammed packing fraction varies by less than one per cent over a wide range of size ratios and stochiometries in amorphous packings of binary and ternary hard-sphere mixtures. We also search for the competing crystalline phases in binary and ternary hard-sphere mixtures over the same range of size ratios and stochiometries using novel genetic algorithms, Monte Carlo and molecular dynamics simulations. We find that the packing fraction of the dense crystalline structures is a complex and non-monotonic function of the size ratio and stoichiometry. We will determine the extent to which the packing fraction and vibrational entropy of the crystalline structures influence the GFA.

Different matter usually has different densities, which is closely relevant to their atomic packaging. Packing problem is a long-standing scientific challenge and has relevance in a wide range of scientific and practical fields.1 Metallic glass (MG)2 with non-directional metallic bonding shows exciting potential for applications and also provides a realistic model system for random sphere packing. Efficient packing (maximizing density) of atoms and/or clusters can stabilize a system and has been revealed to be the basic of mysterious atomic structure in MGs.3,4 Thus the density as a key constrain for modeling plays a essential role to understand the atomic structure and glass forming ability of MGs.4,5 By tuning the density of a MG using integrated multiple in situ high pressure techniques including x-ray diffraction, ultrasonic sound velocity measurements and full field nano x-ray transmission microscopy, we rigorously established a power-law relationship with a power D ~ 2.51 scaling the global density with the atomic level structural information (principal diffraction peak position, 1/q1) in MGs. This study with high quality data provides compelling evidence of the “universal” non-third power-law scaling in MGs, which will put an end to the intense debate in this field.6 The unusual 2.51 power-law oversets our anticipation of a third power-law based on our common sense of random packing in Euclidean space. Our results will have important implications not only in the practical measurements of density or any those involving a change in length scale under various environments, but also in understanding the real atomic packing in MGs and even more general packing problems.
References:
1 C. Song, P. Wang, and H. A. Makse, Nature 453 (7195), 629 (2008); F. Zamponi, Nature 453 (7195), 606 (2008).
2 A. L. Greer and E. Ma, MRS Bull. 32 (8), 611 (2007); W. H. Wang, Adv. Mater. 21 (45), 4524 (2009).
3 J. D. Bernal, Nature 185 (4706), 68 (1960).
4 D. B. Miracle, Nature Mater. 3 (10), 697 (2004); H. W. Sheng, W. K. Luo, F. M. Alamgir et al., Nature 439 (7075), 419 (2006); D. B. Miracle, Acta Mater. 54 (16), 4317 (2006).
5 Y. Li, Q. Guo, J. A. Kalb et al., Science 322 (5909), 1816 (2008); L. Yang, G. Q. Guo, L. Y. Chen et al., Phys. Rev. Lett. 109 (10), 105502 (2012).
6 D. Ma, A. D. Stoica, and X. L. Wang, Nature Mater. 8, 30 (2009); Y. Q. Cheng and E. Ma, Prog. Mater. Sci. 56 (4), 379 (2011); A. R. Yavari, A. Le Moulec, A. Inoue et al., Acta Mater. 53 (6), 1611 (2005); P. Chirawatkul, A. Zeidler, P. S. Salmon et al., Phys. Rev. B 83 (1), 014203 (2011); O. F. Yagafarov, Y. Katayama, V. V. Brazhkin et al., Phys. Rev. B 86 (17), 174103 (2012); N. Mattern, M. Stoica, G. Vaughan et al., Acta Mater. 60 (2), 517 (2012).

Multicomponent bulk metallic glasses are among the most unique and fascinating of the recently discovered materials. However, the lack of thermal stability, arising from their metastable nature and the high tendency of crystallization, has been one of the paramount obstacles that hinders the wide range of applications of metallic glasses. In this talk we will report that the stability of a metallic glass can be dramatically improved by manipulating the local atomic arrangement from short-range order to medium-range order. The glass transition and crystallization temperatures of the ultrastable glass can be increased by above 50 K and 210 K, respectively. The ultrastable metallic glass also shows ultrahigh strength and hardness, over 30 % higher than its ordinary counterpart.

Metallic glasses respond to radiation in qualitatively different ways than crystalline solids. We identify two distinctive mechanisms of radiation response in metallic glasses through a series of ½ billion-atom simulations using molecular dynamics. In the first, inter-nuclear scattering causes localized melting and quenching at rates approaching 10^14 K/s, giving rise to nanoscale “super-quenched zones” (SQZs) with exceptionally high free volume. In the second, rapid volumetric expansion in regions of localized melting generates intense stress pulses that cause polarized plastic deformation in adjacent material. Based on these insights, we construct a parameter for predicting the radiation-response of amorphous materials that may be used in the selection of metallic glasses for applications ranging from nuclear waste storage to ion beam materials modification.

It is considered as the nature behavior that glass deformation involve "disordering" of the structure, or the accumulation of free volume. The dominance of disordering processes is what renders the mechanical properties of glasses (metallic glasses in particular) disappointing-especially properties like toughness and ductility; this general feature of glass deformation is what makes them brittle. In this talk we explore the possibility of deformation-induced ordering in glass. Through detailed experimental work, we show that this unusual kind of deformation can occur in metallic glass at room temperature. We also discuss the physical conditions required to evoke such a process, and implications of these conditions for glass design.

Bulk Metallic Glass, BMGs, -as an engineering metal- is currently an area of growing interest and has focused intense research activities. Among BMGs Fe-based metallic glasses are of great interest because of their attractive soft magnetic properties with coercivity Hc less than 1 A/m, and high saturation magnetization (Ms) values. Furthermore, BMGs exhibit the best mechanical properties, which are of interest to exploit the mechanical and soft magnetic properties of these materials in the thin film form for developing special electromagnetic components and applications in the MEMS technology. We have fabricated by pulsed laser deposition very thin (5-7nm), and thick (27 to 408 nm) glassy films of composition Fe66Ni6B24Nb4 on glass and silicon substrates respectively, and studied their magnetic and magneto-optic properties at room temperature.
We find that the thicker films on silicon can be tuned by appropriate thermal annealing to exhibit soft magnetic characteristics with low Hc, and high Ms values. The magnetic hysteretic loops of the as deposited thicker films on the silicon substrates show two interesting characteristics: 1) increase in the coercivity with the film thickness, and 2) the onset off a two stage process during the approach to magnetic saturation: The initial in-plane characteristic of the hysteresis loop is followed by a linear anisotropic behavior between remanence and saturation- that changes into square soft magnetic loops on decreasing the film thickness. By suitable annealing the intrinsic strain disappears at relatively low temperatures (le;200oC), the thicker films can be tailored to exhibit a simple soft magnetic square loop with low Hc.
The 5-7nm films deposited on glass are transparent and have been investigated for their magneto-optic properties using Faraday rotation measurement technique. Very high values of FR in the range 16 to 20 deg/mu;m film thickness almost linearly dependent on the wavelengths of light in the range 405 - 611nm is observed. The observed high values of Faraday rotation over a wide range of the wavelength of light are useful for the applications as magneto-optic sensors in the UV to visible range.
The studies described above at room temperature on the PLD deposited films based on Fe-based films has been made possible by using XRD, Dual beam SEM/FIB Nova 600 Nanolab, the Faraday rotation spectrometer, and room temperature vibration sample magnetometer. Preliminary room temperature domain observations carried out using Asylum research AFM and MFM system will be also discussed. These soft magnetic transparent BMG based films have then been used to develop non-invasive sensors.
* Hero-M Vinn Center of Excellence supported project.
Carl Tryggers Foundation

Based on the proposed long-range n-body potential, both molecular dynamics and Monte Carlo simulations were carried out to study the ternary metallic glass formation in some bulk metallic glass forming systems containing Cu, Zr, Hf, Al and Ni, etc. In the simulations, solid solution models were employed to compare the relative stability of the solid solution versus its competing disordered state i.e. metallic glass phase, over the entire composition triangle. The simulation results not only clarified that the underlying physics of the metallic glass formation is the crystalline lattice collapsing when the solute concentration exceeds the critical solid solubility, but also predicted, for each system, a quantitative composition region, within which formation of metallic glasses is energetically favored. Furthermore, the energy difference between the initial solid solution and resulting metallic glassy phase was also derived and defined as the driving force for the crystal-to-amorphous transformation. The amount of the driving force could thus be considered as a comparative measure of the glass-forming ability. In a ternary metal system, the largest driving force could be correlated to the optimized composition, of which the metallic glass is the most stable one and likely the easiest one to be obtained in experiments. Interestingly, the predictions directly from the interatomic potentials are well compatible with the experimental observations reported so far in the literature. Consequently, the above described atomistic theory is capable of predicting the glass-forming region for the ternary metal systems and could thus be a good guidance for designing appropriate alloy compositions to produce desired ternary metallic glasses.
Key words: Metallic glass, Glass-forming ability, Atomistic theory, N-body potential
References
1. Jan Schroers: Physics Today 66, 32-41 (2013).
2. J.H. Li, Y. Dai, X.D. Dai: Intermetallics 31, 292-320 (2012).
3. J.H. Li, S.Z. Zhao, et al. J. App. Phys. 109, 113538 (2011).
4. Y.Y. Cui, J.H. Li, et al. J. Phys. Chem. B 115, 4703-4708 (2011).
5. Y.Y. Cui, J.H. Li, et al. App. Phys. Lett. 99, 011911 (2011).
6. S.Z. Zhao, J.H. Li and B.X. Liu: J. Mater. Res. 26, 2049-2064(2011).

From the experimental point of view, it is generally accepted that a material&’s glass-forming ability (GFA) is usually either proportional to its critical casting thickness or inversely proportional to its critical cooling rate. However, for a metal system, differing from the situation of a specific alloy, the Glass-Forming Ability, which is one of the important scientific issues concerning the bulk metallic glass, is quantitatively denoted by the Glass-Forming Range (GFR). The GFR not only shows whether or not metallic glasses could be obtained in a system, but also indicates the alloy composition range, within which metallic glasses could be formed by some specific glass-producing techniques. It follows that the wider the GFR, the greater the GFA of a metal system. To present authors&’ view, a metal system should have its own intrinsic GFA/GFR, as the existence of the atomic configuration in a disordered state is determined by the internal characteristics of the system. The intrinsic GFA/GFR reflects the maximum possible composition range energetically favored for the metallic glass formation. We report in this paper, starting with the interatomic potentials of eight representative binary metal systems, covering various structural combinations and thermodynamics characteristics, molecular dynamics simulations using solid solution models led to establish an atomistic theory for the binary metallic glass formation. The theory not only clarifies that the physical origin of the binary metallic glass formation is the crystalline lattice collapsing while solute concentration exceeds the critical value, but also quantitatively determines, for each system, an exact composition range, within which the disordered state, i.e. formation of metallic glasses, is energetically favored. It turns out that the developed atomistic theory is well supported by the experimental observations reported so far in the literature.

Although some theoretical methods to generate amorphous porous structures have been devised, they are based on either experimental data or experimental procedures which lack transferability to other systems. We use our ab initio approach to generate amorphous porous Cu₆₄Zr₃₆ (ap-Cu₆₄Zr₃₆): the expanding lattice (EL) method, which has been employed previously to generate amorphous porous carbon, silicon, copper, silver and gold. In our approach the interatomic distances are proportionally increased by doubling the volume of the cell, thus halving the density and making the initial supercell metastable. We applied the EL method to two different initial configurations: a 108-atom crystalline cubic supercell (c-Cu₆₄Zr₃₆) and a 108-atom amorphous supercell (a-Cu₆₄Zr₃₆), both with an initial density of 8.06 g/cm3. After the lattice expansion, the initial supercells were subject to ab initio molecular dynamics for 500 steps at constant room temperature using a time step of 10.71 fs. Finally, the resulting structures were relaxed. Pores appeared along well-defined spatial directions. We characterized the structures by means of the pair distribution functions and the bond-angle distributions. Our results were compared with the non-porous amorphous Cu₆₄Zr₃₆. Due to certain difficulties that arise when obtaining microporous amorphous alloys experimentally and the lack of experimental counterparts for ap-Cu₆₄Zr₃₆, our results should be considered predictive. Further studies should be made in order to handle a larger number of atoms, thereby larger pores.

Metallic glasses (MGs) are regarded as strong but intrinsically brittle macroscopically, exhibiting zero tensile plasticity, due to the plastic deformation being localized in a narrow shear band. By introducing a secondary soft phase such as a body-centred cubic β dendritic phase to inhibit and multiple the shear bands, the β-dendrite-reinforced MG composite (MGC) exhibits inhomogeneous elongation of the order of a few percent. Once it yields, its stress decreases continuously, leading to remarkable necking. However, when a β-dendrite-reinforced Ti-based MGC plate is confined by a commercially pure titanium (CPT) substrate, uniform tensile plastic deformation can be achieved in the MGC where the strain localization and necking is suppressed. The MGC can sustain a true tensile strain exceeding 10% without fracture. A deformation mechanism is proposed to interpret the plastic deformation of the MGC/CPT structure. The revealed significant intrinsic plasticity of the MGC suggests its potential application as an excellent coating for bulk ductile materials.

High Entropy Alloys (HEAs), a novel class of material, were developed early this century in the wake of bulk metallic glass development. They are defined as alloys composed of five or more principal elements in equimolar ratios. Similar to bulk metallic glasses in many ways (including production methods), HEAs differ by displaying a crystalline structure which often takes the form of a disordered homogeneous solid solution of face centered cubic (FCC) and/or body centered cubic (BCC) packing. They display enhanced yield strengths and hardness when compared to conventional alloys at temperatures exceeding 1270K (similar to intermetallic compounds), whilst maintaining their ductility. Empirically derived thermodynamic concepts have been used to predict novel multicomponent bulk metallic glass systems. Ab-inito methods were then used to gain a mechanistic understanding of previously studied HEAs as well as to make predictions for new HEA systems. Selected four element samples of these alloys have been produced by vapour deposition techniques. XRD and electron microscopy characterization of the produced phases was carried out to guide the modeling.

The dynamic loss modulus of bulk metallic glasses below the glass transition
temperature is calculated using a thermal activation theory that exploits
the known properties of the potential energy landscape of the under-cooled
liquid regime. Below the glass transition temperature, the calculated loss
modulus is found to consist of two dominant components, which can be
interpreted as reflecting the underlying alpha-relaxation and beta-relaxation
processes that facilitate both micro- and macro-plasticity in these materials.
These results are discussed in terms of recent dynamical mechanical spectroscopy
experiments for a variety of well known bulk mechanical glasses.

A crystalline-to-amorphous-to-crystalline (C-A-C) transition in Cr₂Ti intermetallic compound was investigated by high voltage electron microscopy (HVEM). The Cr₂Ti could not maintain its original crystal structure under electron irradiation, resulting in the irradiation-induced solid-state amorphization (SSA) at and below 103 K. With continued irradiation, a new non-equilibrium crystalline phase was formed from the amorphous phase. Namely, a C-A-C transition was observed under electron irradiation. The crystalline phase obtained through the C-A-C transition was identified as a bcc solid solution phase whose composition was the same as that of Cr₂Ti. A typical C-A-C transition, characterized by an amorphous single phase formation through SSA and the appearance of coarse crystalline grains without the difference in the chemical composition among Cr₂Ti, an amorphous phase and a bcc solid solution, was found. The origin of the C-A-C transition was discussed on the basis of a binary phase diagram and a Gibbs free energy model.

Previous studies have identified compositional medium range order (CMRO) and icosahedral short-range order (ISRO) in phase separating amorphous metal alloys. Using molecular dynamics (MD) simulations of a model phase separating amorphous metal alloy, Cu50Nb50, we show that a percolating network of ISRO forms at interfaces between compositionally enriched regions and leads to glass transition. Below the glass transition temperature, the ISRO network is mechanically stiff, imparts shear resistance, and halts coarsening of the CMRO. The ISRO network constrains the dynamics of surrounding atoms and leads to anomalous diffusion. This ISRO network and its influence on the physical properties of the system bears striking resemblance to gelation in colloidal systems, in which a system-spanning, dynamically arrested network of locally preferred structures imparts stiffness. We discuss how gelation may be relevant to glass transition in more conventional bulk metallic glasses.

The random packing of constituent atoms in metallic glasses (MGs) yields a unique combination of properties. Their excellent mechanical properties, high chemical stability, facile thermal formation, and metallic appearance qualify them as durable, scratch resistant, and aesthetically pleasing touch surface materials. Touch surfaces in public settings serve as reservoirs of pathogenic microbes, which can result in the transmission of infectious diseases. Current practices to prevent bacterial growth on touch surfaces include the use of disinfectants, which requires repetitive operations, and Cu alloys with their intrinsic antimicrobial properties. The efficiency of Cu alloys though can be hindered by the high cost of pure copper, its low strength and wear resistance, and its susceptibility to corrosion. MGs exhibit potential antibacterial property, owing to the presence of biocidal elements (i.e., Cu, Ag, Zn, etc) in various glass-forming compositions. In the current research, the antimicrobial ability of Cu- or Ag-bearing Zr-based MGs was investigated using dry contact assays with the gram positive bacterium Staphylococcus aureus (ATCC 6538). Number of viable adherent cells was determined by serial dilution and the spread plate method and pure Cu or Ag samples were used as controls. A significantly lower number of viable bacterial cells were found on the Zr-Al-Ni-Cu MGs after 4 hr of incubation, indicating the killing capability of the Zr-based MGs. In order to advance the antimicrobial properties of as-cast MGs, the ion implantation technique was employed to increase the surface concentration of biocidal elements. Moreover, we found that the antibacterial mechanisms of Cu or Ag lie in their ion forms. Therefore, the amounts of Cu or Ag ions released from MGs and reference samples were monitored using inductively-coupled plasma (ICP-OES). Ion release profile was established as a function of immersion time, which was directly correlated with killing efficacy.

Amorphous silicon (a-Si) created by self-ion implantation is an extremely pure and reproducible form of amorphous silicon. It is 1.8% less dense than crystalline Si [1] and has no voids [2]. The Si atoms are on average sightly under 4-fold coordinated, due to atomic defects [3]. Continuous a-Si layers created by ion implantation undergo structural relaxation when annealed at low temperatures. This structural relaxation results in a reduction in average bond-angle distortion [4], consistent with a decrease in defects [3]. The radial distribution functions (RDFs) of as-implanted and relaxed a-Si have been measured previously using x-ray diffraction [5]. In that x-ray study the RDFs were almost indistinguishable, except for a 2% increase in co-ordination number upon annealing, consistent with an annealing of volume defects.
Ion-implanted a-Si has also been studied extensively using fluctuation electron microscopy (FEM) [6], a technique that is sensitive to four-body atomic correlations. Through FEM, a model of as-implanted a-Si as an amalgam of strained, topologically crystalline grains approximately 1 nm in size, potentially dispersed within a more disordered, continuous random network (CRN) has emerged [7,8]. These paracrystallite models produce RDFs essentially the same as those of CRNs and both models fit experiment. In this contribution we will present high-resolution RDFs of as-implanted and relaxed a-Si measured using tilted-beam electron diffraction in the transmission electron microscope (TEM) [9]. Diffraction in the TEM has many advantages, such as using dark-field imaging to ensure that there is no contribution from the crystalline substrate. We use our data to constrain models by reverse Monte Carlo, and examine our models' consistency with other observations in the TEM such as aberration-corrected high-resolution imaging and scanning electron nano-diffraction.
[1] J. S. Custer, M. O. Thompson, D. C. Jacobson, J. M. Poate, S. Roorda, and W. C. Sinke, Appl. Phys. Lett., 64, 437-439 (1994).
[2] D. L. Williamson, S. Roorda, M. Chicoine, R. Tabti, P. A. Stolk, S. Acco, and F. W. Saris, Appl. Phys. Lett., 67, 226-228 (1995).
[3] S. Roorda, R. A. Hakvoort, A. van Veen, P. A. Stolk, and F. W. Saris, J. Appl. Phys., 72, 5145-5152 (1992).
[4] S. Roorda, J. M. Poate, D. C. Jacobson, B. S. Dennis, S. Dierker, and W. Sinke, Appl. Phys. Lett., 56, 2097-2099 (1990).
[5] K. Laaziri, S. Kycia, S. Roorda, M. Chicoine, J. L. Robertson, J. Wang, and S. C. Moss, Phys. Rev. Lett., 82, 3460-3463 (1999).
[6] B. Haberl, A. C. Y. Liu, J. E. Bradby, S. Ruffell, J. S. Williams, and P. Munroe, Phys. Rev. B, 79, 155209 (2009).
[7] K. B. Borisenko, B. Haberl, A. C. Liu, Y. Chen, G. Li, J. S. Williams, J. E. Bradby, D. J. Cockayne, and M. M. J. Treacy, Acta Mater., 60, 359375 (2012).
[8] M. M. J. Treacy and K. B. Borisenko, Science, 335, 950-953 (2012).
[9] A. Gorecki, A. C. Y. Liu, and T. C. Petersen, manuscript in review.

A novel experimental strategy for assessing and exploring glass forming ability (GFA) in multicomponent alloys has been developed and applied to ternary, quaternary, and quinary Ni-based metal-metalloid glasses. It is demonstrated that GFA is a rapidly varying exponential function of composition change over the composition regions where nucleation of a given crystalline phase controls glass formation. The overall GFA-composition map is a comprised of these exponential subsurfaces each associated with particular crystallization pathway. Overall GFA is optimized at sharp "exponential hyper-cusps". For Ni-Cr-Nb-P-B glsses, our method leads to novel bulk glasses having critical casting thicknesses exceeding 1 cm. Based on this work, we believe that a broad reassessment of GFA in many well-known alloys is likely to pay substantial dividends.

The development of bulk metallic glasses has been accomplished predominantly through time consuming trial and error approaches. Such a sequential approach is unsuited when considering that bulk metallic glasses are typically based on three or more elements hence vast numbers of possible BMG forming alloys. Here, we introduce a combinatorial approach for developing Au-based BMGs. We use co-sputtering from 3+1 targets to create compositional libraries. The libraries are characterized in a massively parallel manner to identify liquidus temperature, formability, and structural information. This strategy allows us to rapidly identify BMG forming compositions with optimized thermo plastic formability among the thousands of potential compositions.

To date, the design of metallic glass alloys with high glass-forming ability has been accomplished primarily through trial and error, guided by simple empirical guidelines such as critical atomic size ratios or the presence of a deep eutectic. While useful in many cases, these guidelines have been shown to be neither necessary nor sufficient conditions to predict glass formation. As a result, identification of bulk glass-formers in multi-component systems is a time-consuming endeavor, prompting the development of combinatorial approaches to design. In this work, binary, ternary, and quaternary composition libraries were rapidly produced via a laser deposition process. Unlike combinatorial approaches involving sputtering or vapor deposition, the laser deposition process produces pools of liquid with dimensions on the order of hundreds of microns, which quench at rates approaching 10,000 K/s due to the thermal mass of the substrate. The resulting structure of glassy deposits is therefore expected to be more representative of the structure formed via conventional casting or melt spinning. By continuously changing the ratio of the metal powders added to the melt pool, compositionally graded layers ~ 200 microns thick are produced. The composition profiles of the graded deposits have been quantified by energy dispersive spectroscopy for several sets of processing conditions. The featureless surface topography of potential glass-forming regions in these compositionally graded deposits are easily identified by optical microscopy for further characterization. The structure, glass transition, and crystallization behavior of the deposits have been investigated by electron diffraction and differential scanning calorimetry. The mechanical behavior as a function of composition has also been investigated via nanoindentation. Efforts to develop processing maps using thermal data captured in situ during deposition will also be discussed.

Identification of multi-component alloys out of the vast compositional space is a daunting task, especially for bulk metallic glasses that are typically composed of three or more elements. Despite an increased theoretical understanding of glass formation, bulk metallic glasses are predominantly developed through a sequential time consuming trial and error approach. Computational simulations of glass formation, in particular when quantum mechanical methods are used, are by many orders of magnitude short to directly simulate glass formation. Here we present a combinatorial strategy, on the glass forming system Mg-Cu-Y as an example, where ~3,000 compositions are fabricated simultaneously and characterized in parallel. We determine their thermoplastic formability through blow forming the compositional library and rapidly identify compositions with the highest formability. The effectiveness of our method demonstrates drastic increase in the discovery rate of metallic glasses and provides a versatile toolbox for identifying complete correlation, hence better understanding the complex phenomena of glass formation.

To synthesize bulk metallic glasses using low cooling rates was achieved in various alloy systems which possess an extremely high thermal stability before crystallization in undercooled liquid state. In these alloys, the cooling rate is relatively low such that diffusion among the atoms is limited and the amorphization occurs uniformly during solidification. Thus, it is possible to measure the complete Time-Temperature-Transformation (TTT) diagram for the crystallization of the glass forming alloy for the whole range of the undercooled liquid from the melting point down to the glass transition temperature. In the present study, we investigated the metallic glass forming alloys with radiative cooling and experimentally measured the TTT diagrams, using a containerless high-temperature/high-vacuum electrostatic levitation technique. For the crystallization studies, we carefully analysed the microstructure and thermal analysis of the alloy specimens collected after cooling using high energy-XRD , electron microscopes and DSC, which allows us to examine the undercooled liquid stability and phase transformation depending on alloy compositions. Indeed, the experimental condition, where there is no container, can avoid contamination from the wall, so crystallization behaviour can be described by combining steady state nucleation and diffusion controlled growth according to classic nucleation theory. Furthermore, formation of crystalline and metallic glass composite is possible by composition ratio variation. The TTT diagram is the base map showing the material&’s thermophysical property, so critical cooling rate acquired by the TTT diagram could be used to evaluate the GFA systematically and help discover a new metallic glass forming system with high glass-forming ability.

Bulk metallic glasses (BMGs) show many unique physical and mechanical properties such as high strength, relatively low Young&’s modulus and perfectly elastic behavior. However, BMGs undergo failure without sustainable strain after yielding. To solve this problem, various concepts of developing composite microstructure with in-situ/ex-situ formed secondary precipitates have been actively studied. While several Zr and Ti based BMG matrix composites (BMGMCs) showed even tensile ductility, most of these BMGMCs exhibit work-softening behavior with early onset of necking. Recently, Work-hardenable BMGMC system has been developed which contains CuZr shape memory alloy as a secondary phase. But there is no sufficient understanding about the shape memory effect of secondary phase and hardening mechanism. In the present study, we fabricated Ti-based BMGMCs that have an in-situ formed shape memory secondary phase precipitated by polymorphic transformation during solidification. We could control mechanical properties and the phase transformation property of the secondary phase. And we obtained various volume fractions of secondary phase by controlling the amount of additional elements. The composites exhibit large plasticity with work-hardening behavior during deformation. Our result indicates that we can tailor the mechanical properties of the BMGMCs by controlling the phase stability and shape memory effect of the secondary precipitates, and that macroscopic work hardening behavior of these Ti-based BMGMCs can be caused by mechanical and thermodynamical properties of secondary precipitates which can significantly enhance their toughness and sustainability.

Bulk metallic glasses (BMGs) represent a new class of metallic alloys with a wide range of potential applications. However, a major disadvantage of BMGs is the potential for rapid strain localization leading to catastrophic failure, particularly under tensile loading conditions. Among the investigated solutions to this problem is the introduction of a crystalline phase to control the propagation of shear bands through the glassy matrix. In this work, we have studied the compositional profiles across the glass/crystalline dendrite interface of an in situ BMG matrix composite with atom-probe tomography (APT) and analytical scanning transmission electron microscopy (STEM). We find very sharp interfaces with an abrupt change in composition between the dendrites and the glass matrix along with longer-range concentration gradients from the nucleation and growth of the crystalline dendrites during heat treatment. A corresponding model with realistic interfaces was then studied with classical molecular dynamics simulations. Potentials from the Embedded-Atom Method (EAM) are used for model systems and Modified EAM (MEAM) potentials are used for the interfaces, where candidate structures based on the experimental characterization data are proposed for the BCC dendritic structure. We also study the deformation of the composite system under different applied stress states, crystal orientations with respect to the interface, and crystalline volume fractions. For this analysis, the shear strain and hydrostatic strain are calculated for each atom, and variations in these values are compared with the local atomic coordination as a function of time. Following previous work, which suggested a correlation between icosahedral coordination fraction and strain localization, finally we examine the correlation between local strain and atomic-environment.

Mg alloys have long been a subject of interest due to their very low specific weight. However, their wide-spread use is hindered by the relatively low mechanical properties and low corrosion resistance. Manufacturing amorphous Mg-based alloys is expected to improve both the corrosion resistance and the mechanical properties. In the present work the effect of compositional changes on the glass forming ability of Mg alloys containing Y was studied. Five chemically-related rapidly-solidified Mg-alloys were investigated: Mg₉₁Y_7.5La_1.5, Mg₈₅Y₁₂La₃, Mg₈₆Y_9.5Cu_2.5La₂, Mg₈₂Y₁₁La₄Eu₃ and Mg₆₅Cu₂₅Y₁₀. XRD and DSC spectra revealed that the order of the alloys based on descending glass forming ability is the following: Mg₆₅Cu₂₅Y₁₀, Mg₈₆Y_9.5Cu_2.5La₂, Mg₈₅Y₁₂La₃, Mg₉₁Y_7.5La_1.5, Mg₈₂Y₁₁La₄Eu₃. A model providing qualitative and quantitative explanation for the variation in glass forming ability was developed. The model predicts the hierarchy in glass forming ability via thermodynamics of a supercooled liquid alloy and its spinodal-like decomposition.

The shear rate and temperature dependent viscous behavior of Zr_59.3Cu_28.8Al_10.4Nb_1.5 (at%) was measured above the liquidus temperature using a custom built concentric cylinder rheometer. One version of the alloy was prepared using high purity elements and a second version using an industrial grade Zr-Nb pre-alloy. The critical casting thickness for the industrial grade alloy is 5 mm, making this composition attractive for industrial micro-casting applications due to its low material costs. Accurate viscosity characterization is useful not only for defining processing parameters (e.g. for casting and thermoplastic deformation) but also for investigations of kinetic fragilities. Changes in fragility may identify possible liquid-liquid transitions such as those proposed for Zr_41.2Ti_13.8Cu_12.5Ni_10.0Be_22.5 (Vitreloytrade;1).
The apparatus was designed to operate at temperatures up to 1373 K, to apply shear rates up to 1000 s^-1, and to have a viscosity measurement range from 1-1000 mPa s. In one set of experiments, shear rates were varied using a stepwise profile and the temperature was held constant. A second set of experiments involved shearing at a constant rate while applying various thermal treatments, such as cooling from above the liquidus temperature until the onset of crystallization. Additional isothermal three point beam bending experiments were conducted near the glass transition temperature in a thermomechanical analyzer. Vogel-Fulcher-Tammann fits of all of the viscosities are compared to determine the impact of impurities on the kinetic fragility of the alloy.

The dynamics in fluids is governed by two intimately related properties: viscosity and atomic diffusion. While the first describes macroscopic transport of momentum by collective motion of the particles, the latter describes single particle diffusive transport. A common relation, that is often taken for granted in order to calculate required diffusion coefficients of atoms or molecules in a liquid from the viscosity, or vice versa, is the Stokes-Einstein relation. This relation was derived in order to study the diffusive motion of a mesoscopic sphere in a viscous medium. However, when the diffusing objects are of atomistic size, deviations of the diffusion coefficient and the viscosity from the Stokes-Einstein behaviour can be observed.
From an experimental point of view, a direct proof of the Stokes-Einstein relation is still very difficult due to the lack of reliable experimental data of the transport coefficients. The accurate measurement of diffusion data, using long capillary methods, is subject to large errors due to additional transport of mass by buoyancy driven flow effects. Pollution of the sample from chemical reactions with the container walls complicates the measurement of both properties. In order to check the Stokes-Einstein relation measurements need to be carried out over a sufficiently large temperature range. This includes high temperatures favouring convection and chemical reactivity. With the development of advanced containerless processing techniques, such as electrostatic levitation (ESL), we are now in a position to master these challenges.
Using ESL, we carried out quasielastic neutron scattering (QNS) experiments and determined accurate self-diffusion coefficients, in liquid Zr-Ni and bulk glass-forming Zr-Ti-Ni-Cu-Be alloys. These data were measured over a broad temperature range from 1000 K to 1700 K. Viscosity of these liquids were also measured via the oscillating drop technique in ESL over a broad temperature range: With these results for a dense glass forming system, the relation of viscous flow and diffusion of mass could be checked in unequalled detail [1].
Our data show that the product of diffusion and viscosity is constant for the entire temperature range contradicting the Stokes-Einstein relation. According to Mode-Coupling-Theory (MCT), the dynamics in a liquid is strongly coupled when the particle density is large. This leads to a freezing of the atomic motion when MCT&’s critical temperature Tc is approached upon cooling. In this case, the diffusion coefficient and the inverse of the viscosity are proportional to the same scaling law. Our results have been obtained at temperatures well above Tc. Apparently possible deviations from the MCT scaling behaviour are similar with respect to temperature for both mass transport coefficients.
[1] J. Brillo, A. I. Pommrich, A. Meyer, Phys. Rev. Lett. 107, 165902 (2011).

Polymorphic transitions are common in crystalline solids. In the last decades, experimental and simulation studies suggest that there exist liquid-liquid transitions (LLT) between two liquid forms with the same composition but different structures and properties. The examples include water, Al₂O₃-Y₂O₃, SiO₂, BeF₂, molten phosphorus (P) and also the binary metallic glasses Ce-Al, although it is a long-standing debate regarding the nature of LLT and even its existence due to the difficulties of experimental observations. Recent viscosity measurements imply the existence of LLT transition in the bulk metallic glass-forming liquids.
In this work, we carried out structural investigations of the strong-fragile transition on Zr-based bulk metallic glass-forming systems during cooling and heating using the in-situ synchrotron X-ray scattering experiments in contactless environment with an electrostatic levitator (ESL). The structural evolution of liquid is observed from the molten liquid state down to the glassy state without crystallization. We observe anomalous structural changes in the liquid state. These local structural changes are consistent with the strong-fragile transition observed in viscosity measurements and can be considered as the structural origin of the anomalous thermal events. The observed local structural changes are suggestive of a liquid-liquid transition in the system and may explain the dynamic crossovers (strong-fragile transition) and the thermodynamic anomalies.
The transition behavior in the metallic liquids is discussed as an underlying lambda transition in liquids. It is compared with other strong glass-forming liquids, such as SiO₂ and BeF₂ in which a LLT in the stable liquid state is suggested by numerical simulations. The LLT transition could be a general feature of strong glass-forming systems and is related to liquid and glass properties which are also of importance for applications.

We present a method to join metallic glasses in air. The joining is achieved by thermoplastic deforming. A model is developed which quantifies the bonding strength solely from the shear strength of the metallic glass and the applied strain. Mechanistically, during straining of the interface, the oxide layer breaks and pristine alloy flows towards the interface and forms a metallic bond. The incipient surface roughness sheds a significant influence on the joined interfacial microstructure, specifically the distribution profile of the joined area (pristine contact) and the residual fractured oxide layer. The residual discontinuous oxide acts as an array of “microcrack”, the average length scale of which increases with the initial surface roughness. The ability to join even reactive metallic glasses such as those based on Zirconium in air within a short time scale on the order of milliseconds at a low pressure and temperature with predictable strength suggest a highly practical and economic joining method.

Nanoporous metals have been produced by dealloying in aqueous solution, by selective corrosion of less-noble component(s) from a multicomponent alloy solid. This dealloying method has been used for the preparation of noble metals such as Ni (well known as Raney nickel catalyst), Cu, Au, Pt, and Pd. Recently, we developed an alternative dealloying method using a metallic melt instead of an aqueous solution. The mechanism can be understood by the mixing enthalpy of constituent elements of the precursor and metallic melt. We have successfully prepared various nanoporous base metals such as α-Ti [Mater. Lett. 65, 1076 (2011)], β-Ti alloy [Scr. Mater. 65, 532 (2011)], and Fe, Cr, and Fe-Cr stainless-steel alloy [Scr. Mater. 68, 723 (2013)]. For example, submicron porous Ti was fabricated by dealloying reaction, i.e., removing Cu element from a Ti-Cu alloy in a Mg melt. By applying this reaction, a Mg-Cu-Gd BMG matrix composite with porous α-Ti dispersoids was successfully prepared by removing Cu element from Cu-Ti-Gd prealloy, which consisted of Ti-Cu and Cu-Gd intermetallic compounds, in a Mg melt [Mater. Sci. Eng. A in press, doi.org/10.1016/j.msea.2013.06.015]. Because the porous Ti dispersoid was imparted from the Ti₂Cu precipitate in the prealloy, size of the dispersoids, thus, mechanical properties of the composite, was controlled by cooling rate of the prealloy solidification. With the dealloying method, a Zr-Cu-Al-Ag BMG matrix composite with Ta dispersoids, which is finer than those obtained with the conventional method, was also prepared [Mater. Trans.54, in press].

Experimental and theoretical studies show that metallic liquids develop significant short- and medium-range topological order with increased supercooling. Icosahedral short-range ordering (ISRO) is typically dominant. This structural ordering can have a strong influence on the crystal nucleation barrier, has been identified as playing an important role in the glass transition, and can correlate with chemical ordering. Liquids are often described in terms of their fragility. They are strong when the viscosity is Arrhenius over a wide temperature range. They are successively more fragile when the viscosity shows a non-Arrhenius (Vogel-Fulcher-Tammann, stretched exponential, etc.) behavior. Our experimental measurements of the density of Cu-Zr liquids as a function of temperature show that the fragility is also reflected in the volume expansion coefficient of the liquid. As would be expected, a correlation is also observed between structural ordering in the liquid and its fragility. As will be shown, the magnitude of the height of the first peaks in S(q) and g(r) and the amount of icosahedral order, determined from Reverse Monte Carlo fits to the measured S(q), all show more abrupt changes on approaching the glass transition temperature for more fragile liquids.
Supported by the National Science Foundation (DMR-08-56199 and DMR-12-06707) and NASA (NNX07AK27G & NNX10AU19G).

Novel phosphorous-free Platinum-based bulk metallic glass has been developed based on the Pt-Si-B system. Increasing substitution of Pt with Cu, was found to considerably improve the Glass forming ability (GFA) (Trg>0.6). Further optimizing the metalloid content led to considerable increase in the width of the super-cooled liquid region (Tx-Tgasymp;70K). Based on Differential Scanning Caloriometry analysis, this increase was mainly due to an increase of Tx while Tg decreased only slightly. The increase of Tx resulted from a shifting of the crystallization peak of nano-Pt to higher temperatures whereas the crystallization temperature of the first complex crystalline phase remained nearly unchanged.
The glassy state was further characterized by viscosity measurements by parallel plate rheometry, for samples with different aspect ratios. The fragility indices of m=30hellip;45 identifies these BMGs as decently strong glass formers.

Nanoporous metals can be produced by either chemical or electrochemical removal of the less noble element/phase of an alloy of suitable composition (i.e. by dealloying). The resulting material is made of a network of ligaments of the most noble element separated by pores. Its properties are of interest in various fields such as catalysis1, surface-enhanced Raman scattering (SERS)2 and actuators3 for which the size of ligaments and pores must be properly tuned. This is achieved by changing etching parameters such as electrolyte, temperature and time.
De-alloying of an amorphous precursor is attractive to get a bulk, self-sustaining porous material with fine microstructure after copious nucleation of crystalline hillocks. However, the noble metal atoms, freed on the alloy surface, gain high mobility causing coarsening of ligaments4. This work describes both the electrochemical and free corrosion de-alloying of Au-based amorphous alloys contrasting them with crystalline solid solutions. Dimension of ligaments and pores have been evaluated by electron microscopy (SEM and TEM). Extensive XRD analyses provided lattice constants, i. e. composition, and grain orientation of the porous material. The data are used to evaluate the surface diffusion coefficient (Ds) and the activation energy (Ea) for diffusion of Au adatoms. Examples of catalytic reactions and SERS will be provided.
1 X.Y. Lang, et al., J.Phys. Chem. C 114 (2010), 2600-2603.
2 G. Li, et al., Solid State Sci 13 (2011), 1379-1384
3 H.-J. Jin, J. Weissmüller, Adv. Eng. Mater. 12 (2010), 714-723.
4. F. Scaglione et al., Journal of Alloys and Compounds, doi.org/10.1016/j.jallcom.2012.11.029.

Bulk metallic glasses (BMGs) possess many of the desired characteristics of electrocatalysts. They exist in a wide range of compositions and can be thermoplastically formed into complex geometries over length scales ranging from 10 nm to a few centimeters [1]. The unique amorphous structure of the BMG produces interesting properties such as very high strength and elasticity combined with good corrosion resistance. Previously, we have shown that Pt57.5Cu14.7Ni5.3P22.5 BMG nanowires are promising electrocatalysts for methanol and ethanol oxidation with high activities and excellent durability in acidic electrolytes [2]. Further studies were shown that the structure of the BMGs can be tuned via dealloying, to produce dendritic or nanoporous structures [3, 4]. While the dealloying procedure removes less noble metals, dealloying alone cannot add elements to the surface of the BMG.
In this this study we will show how free standing Pt42.5Cu27Ni9.5P21 BMG nanowires can be modified with additional elements for specific needs. In our first example, we demonstrate that dealloying the Pt42.5Cu27Ni9.5P21 BMG produces a high surface area nanoporous structure with extreme durability. In fact 86% of the electrochemical active surface area (ECSA) after 3000 cycles is retained, in comparison to Pt/C ETEK, which loses ca. 100% of its ECSA (due to carbon corrosion) after 1500 cycles. In our second example, we show that Ru can be readily deposited onto the surface using either galvanic displacement or underpotential deposition, for enhanced methanol electrooxidation activity. Using this “tool-box” of methods allows for the fabrication of active surfaces containing elements that are outside of the glass formability regime, but that can also take advantage of the BMG freestanding nanowire high surface area architecture.
References:
[1] R.C. Sekol, G. Kumar, M. Carmo, F. Gittleson, N. Hardesty-Dyck, S. Mukherjee, J. Schroers, A.D. Taylor, Small 9 (2012) 2081-2085.
[2] M. Carmo, R.C. Sekol, S.Y. Ding, G. Kumar, J. Schroers, A.D. Taylor, ACS Nano 5 (2011) 2979-2983.
[3] S. Mukherjee, M. Carmo, G. Kumar, R.C. Sekol, A.D. Taylor, J. Schroers, Electrochimica Acta 74 (2012) 145-150.
[4] S. Mukherjee, R.C. Sekol, M. Carmo, E.I. Altman, A.D. Taylor, J. Schroers, Advanced Functional Materials 23 (2013) 2708-2713.

Metallic glasses, also known as amorphous metals, exhibit unique mechanical properties in terms of their strength and ductility. In this work, molecular dynamics simulation techniques were adopted to prepare (Cu50Zr50)100-xAlx thin-film glasses, with x = 0, 2, 4, 5, 6, 8, and 10, were prepared by a simulated sputter deposition process. The deposition simulations were conducted with a tight-binding interatomic potential, and argon working gas was modeled by the pair-wise Moliere potential. Subsequently, the deposited amorphous films were simulated for their nano-indentation properties with a right-angle conical indenter tip. Simulation temperatures were set to temperature 300, 500, 700 and 900 K to compare with experimental data. The pair distribution functions of the films at the testing temperatures were calculated. In addition, the hardness and Young&’s modulus, as well as the pileup index, were calculated from the nanoindentation simulations. Our MD simulation results are compared with experimental data. Furthermore, atomic stress, strain and elastic constants were calculated to reveal deformation localization. Deformation mechanisms in the thin-film glasses are then discussed.

Structural excitations of model Lennard-Jones glass systems are
investigated using the Activation-Relaxation-Technique (ART), which
explores the local potential energy landscape of a local minimum energy
configuration by converging to a nearby saddle-point configuration.
Such information gives the barrier energy corresponding to that particular
saddle-point and when many such barrier energies are obtained and binned,
a single-peaked distribution results for well relaxed samples. The present work
characterizes these atomic scale excitations in terms of their local
structure and environment. This is done for zero and non-zero values of applied strain.
It is found that many of the identified events consist of chain-like excitations which can
either be extended or ring-like in their geometry, and that a low barrier energy correlates
with both the size of the atom and to a lesser extent, the local curvature of the potential energy landscape.

It is well known that applications of bulk metallic glasses (BMGs) for structural materials are limited by their brittleness in spite of desirable mechanical properties such as high yield strength, high elastic limit, and relatively low elastic modulus. To overcome this drawback, various composite microstructures with structural heterogeneity of micrometer to nanometer scale in the amorphous matrix have been developed. The heterogeneity can result in nucleation of multiple shear bands and interruption of their propagation, which can lead to enhanced plasticity. In the present study, we propose a novel BMG composite design concept in which structural heterogeneity is manipulated down to atomic scale by compositional tailoring.
Experimentally, Cu-Zr-Al-Y alloy systems were designed by exploiting the positive heat of mixing relations between Zr and Y. We explored how the degree of heterogeneity responds to changes in the amount of substitution between Zr and Y in Cu-Zr-Al-Y alloy systems. In our experiment, alloys with a large amount of Y over 10 at.% are shown to lead to nanometer-scale phase separation in Cu-Zr-rich and Cu-Y-rich glass phases. On the other hand, alloys with a small amount of Y under 5 at.% generate atomic scale heterogeneity in BMGs. The size distribution and chemical composition of secondary amorphous phases are evaluated by analyzing SANS/SAXS patterns. The thermally-activated change of the heterogeneity is also investigated through in-situ SAXS/WAXS measurement. In connection with the intensive microstructure characterization, the mechanical properties of Cu-Zr-Al-Y alloys have been systematically studied through nanoindentation test and kink angle measurement from bending test. This provides us with a fruitful idea on how to improve plasticity in BMGs with amorphous heterogeneity in nanoscale.

One of the major challenges to overcome to use bulk metallic glasses in structural applications is their apparent brittleness. Rather than an intrinsic feature of the glassy state, this apparent brittleness arises from a strong localization of deformation at low temperatures via formation of shear bands since the deforming regions (shear bands) display large plastic strains themselves. Within the description of the free volume model, deformation is controlled by the amount, distribution, creation and annihilation of excess free volume. Experimental observations show that decreasing the amount of free volume by structural relaxation via annealing results in an embrittlement of the material. However, a rejuvenation of the material, which corresponds to lifting the material to a higher free volume or enthalpic state should manifest in ductilization. Due to the fact that shear-induced irreversible atomic rearrangements are dilatation, plastic deformation should result in rejuvenation reflected by an increase in fracture strain, and a decrease in yield strength and hardness.
In order to investigate how the structural changes associated with rejuvenation via plastic deformation influences the mechanical behavior, 2-D cold rolling of a Zr-based bulk metallic glass was performed. The resulting structural changes were monitored as a function of the true plastic state using X-ray diffraction, Differential Scanning Calorimetry as well as optical and electron microscopy. To complement the topological changes with mechanical behavior, hardness was measured as a function of true plastic strain.
Our work shows that both the shear band density and free volume continuously increased with ongoing deformation. By comparing to published data, we observed a common square-root dependence of the shear band density with respect to true plastic strain independent of the composition or loading mode (cold-rolling, compression). Moreover, the increase in free volume was found to scale linearly with the number of shear bands. Based on an analysis of the measured shear band densities and enthalpy changes, it is concluded that the free volume of both the matrix and shear bands must continuously evolve during deformation. Moreover, the hardness was found to initially decrease, passing a minimum at a true plastic strain of about ~0.07, recapturing its initial value at higher deformation degrees. While the initial decrease could be attributed to the generation of free volume and shear bands, the recovery was related to compressive stresses emerging at high deformation degrees.

Amorphous metals have special interest because of their superior mechanical properties such as high elastic strain limits, high strength and wear resistance. The lack long range translational symmetry in the atomistic arrangement in amorphous metals contributes to mechanical behaviors which are unique and outstanding. Nonetheless, those materials do not exhibit macroscopic plasticity, because of localization of plastic flow in shear band, which nucleate and propagate globally and cause the failure.
Mechanical responses of amorphous Fe₈₀B₂₀ cylindrical nanowires having diameters of 2.0-6.0 nm under tensile load have been investigated using molecular dynamic simulations. Applied strain rates during tensile test lies in range 3.5x109 s-1 and 3.5x108 s-1. Interatomic interaction potentials were calculated from experimental partial pair distribution function via Inverse Molecular Dynamic method. Fe-B amorphous nanowires have been extruded from bulk amorphous alloy in order to acquire intrinsic mechanical properties of samples with smooth surfaces. For sample having 7% plastic deformation, unloading test simulations have also been provided. It has been observed that Fe₈₀B₂₀ nanowire samples are strain hardenable and their mechanical properties are sensitive to strain rate.

MgCaZn metallic glasses are good candidate for biodegradable implants for their biocompatibility, low density, high strength and low modulus comparable to human bone. However, they usually suffer from high corrosion rate in human body due to their low corrosion resistance. Finding new MgCaZn alloy of suitable corrosion rate out of the ternary compositional space are typically time consuming with conventional methods. In the present work, we employ magnetron confocal sputtering to create MgCaZn material library spanning large compositional range (Mg: 30-80 at.%, Ca: 2-32 at.%, and Zn: 10-60 at.%). We demonstrate that we are able to scan the corrosion rate of across the library and quickly locate the composition with best corrosion resistance within the library. Furthermore, microstructural characterizations indicate that the corrosion behavior of the MgCaZn system is strongly composition-dependent and consistent with corrosion rate measurements.

Based on hybrid reverse Monte Carlo structural refinement against fluctuation microscopy data and an empirical interatomic potential, we have proposed that Zr50Cu45Al5 bulk metallic glass (BMG) contains two types of nanometer-scale structure [1]. One type is icosahedral nearest-neighbor clusters arranged in chains, and the other is compact, crystal-like clusters with 4- and 6-fold approximate rotational symmetry. As a function of annealing, crystal-like clusters evolve into icosahedral-like clusters.
Here we report FEM experiments on Zr54Cu38Al8 BMG, which is a poorer glass former, closer to the edge of one of the glass forming composition windows in the Zr-Cu-Al system. We find features in the FEM data that corresponds to the icosahedral-like and crystal-like structures found in Zr50Cu45Al5. In the as-cast state, the crystal-like feature is stronger and the icosahedral-like feature is weaker in Zr54Cu38Al8. Like Zr50Cu45Al5, the icosahedral-like feature grows stronger with annealing time at 0.83Tg (573 K). The stronger crystal-like order in the as-cast state may be associated with the poorer glass forming ability. There is also a new feature in the Zr54Cu38Al8 FEM data at slightly higher scattering vector, the origin of which remains to be explained.
[1] J. Hwang, Y. Kalay, I. Kalay, M. J. Kramer, D. S. Stone, P. M. Voyles, Phys. Rev. Lett. 108, 195505 (2012).

Novel Phosphorous-free Platinum-based bulk metallic glasses have been developed based on the Pt-Si-B ternary system. The influence of substituting part of the Pt, Si, and B by late transition metals and by other metalloids, respectively, on the characteristic parameters for glass formation has been studied. Glassy alloys with critical casting diameter 3mm and a Tx-Tgasymp;70K have been obtained by copper mold suction casting.
The glassy alloys exhibit excellent microhardness in glassy state (540HV). For exploring the influence of postponed heat treatment on glassy alloys, isothermal annealing has been carried out in stabilized salt bath i) at temperatures below the glass transition to see the effect of structural-relaxation-related phenomena on microhardness, and ii) at temperatures above the glass transition temperature introducing partial (nano)crystallization, leading to superior microhardness of up to 850 HV. The concomitant microstructural changes have been investigated by Transmission Electron Microscopy and X-ray diffraction analyses.

The amorphous atomic structure of bulk metallic glasses (BMGs) enables thermoplastic forming, which can be employed to create precise substrates to investigate cell-nanopattern interactions. In this study, we fabricated four distinct nanopatterned platinum-based BMGs with nanorods of 55, 100, 150, and 200 nm in diameter. In addition, the aspect ratio of nanorods (height/diameter) ranged from 6 to 10. Cell-nanopattern interactions were investigated using three cell types involved in the foreign body response and tissue repair , namely, 3T3 fibroblasts , primary macrophages and human umbilical cord endothelial cells (HUVECs). We detected nanopattern induced changes in cell size and shape, which were quantified by cell area/perimeter and cell circularity/elongation factor, respectively. NIH 3T3s became increasingly smaller in size, more circular, and less elongated on the BMGs in the following order - Flat BMG, BMG-55, BMG-100, BMG-150 and BMG-200. This behavior suggested that 3T3 fibroblast cells could detect all nanopatterns studied. Moreover, these changes in cell spreading were mediated by changes in protein complexes involved in cell adhesion (focal adhesions) and compromised activation of Rho-A GTPase, an intracellular regulatory protein involved in cytoskeletal remodeling. In contrast, macrophage morphology differed only on BMG-200 where cells were larger in size, less circular and more elongated. HUVECs displayed similar morphology on flat BMG and BMG-55 suggesting that they cannot detect the smallest nanopattern. These observations indicate that the lower limit of nanopattern detection is cell type dependent. Constitutive linear regression correlations between substrate nanotopography to resultant cellular morphology were also developed. Currently, we are probing changes in cellular traction forces induced by nanopatterns with nanoscale precision by employing focused ion beam scanning electron microscopy. Our ultimate goal is to develop a strategy for rational nanoscale design of BMGs to engineer cellular responses.

Upon heating above the glass transition temperature, a metallic glass often devitrifies into crystalline phases as high-density nanoscale particles. In this work, a consecutive two-step phase-transformation scheme has been proposed to account for the nanocrystallization pathway, based on our neutron diffraction and small-angle neutron scattering (SANS) studies of a multicomponent bulk metallic glass. The neutron scattering experiments were conducted on the NOMAD beamline at the Spallation Neutron Source (SNS), and GPSANS at the High Flux Isotope Reactor (HFIR), Oak Ridge National Laboratory. The complementary use of neutron diffraction and SANS provides structure information over multiple length scales ranging from atomic levels to tens of nanometers during crystallization. The knowledge gained from the present study has implications for better understanding multiscale phase-transformation kinetics of complex amorphous materials.

Titanium has been attracting attention as the next generation of lightweight materials, followed by aluminum and magnesium alloys, because of its high corrosion resistance and specific strength. Especially, beta titanium alloys have been actively studied for high value-added industries, such as biomedical materials and advanced structural materials due to the low modulus of elasticity, high tensile strength, excellent formability, and shape memory characteristics. In case of metastable beta titanium, reversible martensitic phase transformation is available through varying temperatures. The phase transformation temperature can be modified by tailoring alloy composition related to beta stabilization, and it is closely related to stress level of stress-induced martensitic transformation. Based on these theories, the present study develops systems of high-strength titanium alloys by optimizing stress level related to stress-induced martensitic transformation. The alloy systems have unique properties, such as super-elasticity and high strength. Therefore, we will discuss the relationship between mechanical properties and phase stability controlled by alloy composition. This work will be extended to the development of a novel bulk metallic glass matrix composite with a new type transformation media.

Metallic glasses are a class of material with incredible potential due to their outstanding mechanical properties, formability at moderate temperatures, and corrosion resistance. However, their characteristic low tensile ductility currently limits the usage of monolithic metallic glasses for structural applications. Introducing a crystalline phase can produce metallic glass matrix composites with drastically improved toughness and ductility. While some basic rules exist describing the desirable spacing of the crystalline phase and the relative moduli of the phases, the exact relationship between processing conditions, microstructure, and mechanical behavior of metallic glass matrix composites is not well understood. Such understanding is necessary to economically design metallic glass composites with desirable properties. To address this need, we have performed serial sectioning to reconstruct and quantitatively describe the three-dimensional microstructure of selected metallic glass matrix composites reinforced by dendrites precipitated in situ. The elastic and plastic deformation behaviors of the constituent phases have been characterized using nanoindentation, with particular focus on understanding the influence of the glass/crystalline interface. With these data, we have meshed the reconstructed 3D microstructures to construct and analyze a finite element model, with the goal of understanding each phase&’s contribution to the overall composite performance. The modeling results are validated by linking the simulated results to the processing history of the input microstructure, and comparing directly against corresponding experimental measurements of the bulk tensile and fracture behavior of selected composites. In this way, a large number of experimental or hypothetical microstructures can be analyzed efficiently and economically, clarifying the relationship between composite structure and performance.

Fe-based amorphous alloys exhibit excellent soft magnetic properties in both amorphous and partially de-vitrified state. The magnetic properties of these alloys can be improved by de-vitrification of amorphous precursors. In the current work, it is shown that laser annealing of the amorphous ribbon improves the magnetic properties compared to conventional furnace annealing. The phase evolution and compositional analysis after laser material interaction has been studied by multi-modal characterization using micro-XRD, TEM and Atom Probe Tomography. The evolution of microstructure is rationalized using a Finite Element based heat transfer multi-physics model. The improvement in magnetic properties is explained on the basis of microstructure evolution and incomplete solute partitioning of alloying elements.

The recently developed self-learning metabasin escape algorithm is coupled with shear deformation to elucidate strain-rate and temperature dependent shear transformation zone (STZ) characteristics in two-dimensional amorphous solids. We report a transition in the STZ characteristics that occur as the shear strain rates are reduced to those that are experimentally accessible, including a 33% increase in STZ size, faster spatial decay of the displacement field, a change in deformation mechanism inside the STZ from shear to tension, a reduction in the stress needed to nucleate the first STZ, and finally a notable loss in characteristic quadrupolar symmetry of the surrounding elastic matrix that has previously been seen in athermal, quasistatic shear studies of STZs.

A glassy alloy is characterized by its own unique loosely packed random structure and some properties of glassy alloy originate to the unique random structure; i.e. thermoplastic deformation due to glass transition phenomena and inhomogeneous deformation due to no operable dislocation. By the optimization of homogeneity and solidification process of molten alloy, vitrification can be achieved in various scales from centimeter-sized rod to single micrometer spherical powder. In this presentation, I would like to talk about my resent topics;
1.Casting ~Micro Factory Concept for Bulk Glassy alloys~ [1]
Good reproducibility should be required for standardization and industrialization of cast bulk glassy alloy. The Automatic fabrication system, which composed of automatic weighing, alloying and casting, was developed to achieve good reproducibility. Pre-alloying process was also examined to suppress the formation of harmful intermetallic compound during alloying process. Making full use of the automatic fabrication system and prealloying process, meaningful progress will be expected on the advanced quality control of bulk glassy alloy, in addition, the process promises the avoidance of human error and dependence on human skill.
2. Atomizing ~Fe-based glassy alloy fine powders~ [2]
Powder metallurgy is one of suitable processes to mould glassy alloys into industrial products with reasonable price, however, present conventional powder making processes (gas and water atomization) seem difficult to reduce the price of glassy alloy powders. Therefore, new categorized atomization process for Fe-based glassy alloy powders had been tried to develop by using high velocity air fuel flame (HVAF), and the process was named counter flame jet atomization (CFJA). This CFJA, which can be expected to produce small sized powder (i.e. single micro meter in diameter) by its unique atomization conditions of high temperature (~1600°C) and high velocity (~1600 m/s), enables to fabricate Fe73.2Cr2.2Si11.1B10.8C2.7 fine glassy alloy powders with combination of specified cooling system.
[1] http://www.diavac.co.jp/english/products/souti/kogata-a-ku/index.html
[2] http://www.tohoku.ac.jp/english/2013/01/press20130125-01.html

While fabrication of semiconductors and polymers has surged into sub-nanometer regime, metal processing still lacks a scalable method for sub-micron (< 1000 nm) features. Nanolithography and self-assembly are not readily applicable to metals because of their isotropic etch rates, high surface energy, and rapid grain growth. Here, we present a comprehensive strategy to fabricate a variety of nanostructures (nanoparticles, nano-gratings, long nano-wires, ultra-sharp tips, and graded nano-pores) by template-free restructuring of simple patterns (pillars and holes) in metallic glasses. This two-step approach mitigates the challenges of direct nano-patterning which requires expensive templates and optimal wetting properties. In addition, nano-structures with a systematic variation in geometries can be fabricated by using a single template to study the size- and shape-dependent properties. Potential applications of nanostructured metal templates in sensing, bio-detection, self-assembly, and fundamental studies will be discussed.

Metallic glasses being non-equilibrium materials are thermodynamically unstable with respect to the crystal phase. At temperatures above the glass transition, where the crystal phase becomes kinetically favored, the incubation time for crystallization decreases hyper-exponentially with temperature. Incubation times at temperatures high enough into the supercooled liquid region associated with the more mobile liquid states are often on the order of milliseconds or less. Conventional heating methods for metals (e.g. inductive heating or resistive furnace heating) generally supply heat at the surface of a sample, and rely on thermal conduction across the sample to attain a homogeneous state of thermal equilibrium. Typical thermal relaxation times associated with heating macroscopic (i.e. millimeter size) metallic glass samples are on the order of a second, i.e. much greater than typical incubation times at those temperatures. Thermal relaxation therefore severely limits access to those high temperature states. Even unconventional heating methods such as laser heating capable of inducing very high surface heating rates would not address the problem, as they still rely on thermal relaxation to reach thermal equilibrium. Heating therefore should be not only fast, but also uniform, such that it evolves rapidly, stably, and homogeneously (i.e. volumetrically) throughout the material. Such heating is achievable in metallic glasses through the discharge of electrical energy. Owing to the relatively high and nearly temperature-independent resistivity, electrical energy would dissipate uniformly across a metallic glass such that heat would be generated volumetrically. Therefore, discharging a high frequency capacitor would generate rapid uniform heating of the metallic glass and enable access to essentially any temperature in the supercooled liquid region for a finite time prior to crystallizing. Over this isothermal time window, a measurement could be taken or a processing step could be performed. In this talk, scanning calorimetry data through the glass transition and measurement of enthalpy recovery at heating rates orders of magnitude higher than conventional calorimetry achieved by capacitive discharge heating will be presented, along with crystallization kinetics data and TTT diagrams extending to temperatures inaccessible with conventional calorimetry. Furthermore, the ability to process two- or three-dimensional net-shape parts at temperatures where the viscosity is in the range of 10² to 10⁴ Pa-s by forging or injection molding will be demonstrated.

The oxidation behaviour of various metallic glass systems such as Zr-, Ti- and Al-based has been investigated with a special emphasis on the oxidation behaviour in the supercooled liquid region. There is a large difference in the oxidation resistance depending on the alloy chemistry. We suggest that the oxidation resistance of the metallic glasses during heating is dependent on the thermal stability of the amorphous oxide layer. The oxidation resistance of the various metallic glasses is compared on the basis of the thermal stability of the amorphous structure of the oxide layer. The result shows that formation of amorphous oxide layer with enhanced thermal stability can significantly improve the oxidation resistance of metallic glass in the supercooled liquid state. Addition of a small amount of Be (2 at.%) in Cu46Zr46Al8 widens the supercooled liquid region and simultaneously improves the oxidation resistance of the supercooled liquid. We further propose that the stability of the amorphous oxide layer is highly dependent on the interface stability between the amorphous oxide and the amorphous substrate. The interface stability is closely linked to the thermal stability of the amorphous substrate, which is changed by the chemical composition shift during formation of the oxide layer. Therefore, significant improvement of the oxidation resistance in metallic glass systems can be achieved by the control of the compositional changes underneath the oxide layer. The result also shows that the amorphous oxide layer in Al-based metallic glass has a high thermal stability when the oxide layer is doped with the proper alloying element such as Y. The amorphous layer in Al-Ni-Y metallic glass is stable after continuous heating up to 873 K, and grows retaining the amorphous structure when isothermally heated at 873 K. However, long term exposure leads to the formation of crystalline oxide layer at the oxide/matrix interface.

We report on the wettability of different nano-textured metallic glass surfaces. The surface patterns on the metallic glass were obtained by electrochemical processing. Wettability was quantified by the sessile drop technique using distilled water droplet. We demonstrate that the hydrophilic-hydrophobic nature of the metallic glass surface can be controlled with different nano-textures. Atomic force microscopy was used for detailed evaluation of surface topography and roughness. Contact angle was found to be directly correlated to the surface roughness.

Bulk metallic glasses (BMGs) possess excellent mechanical properties such as high strength and elasticity, reveal good wear resistance, and usually corrode homogeneously due to a lack of galvanic corrosion and the absence of grain boundaries. Because of these properties and various processing advantages, BMGs are of great interest for biomedical applications. Here I will focus on Mg-based BMGs for deployment in biodegradable implants, which degrade in the body over time and thus do not require surgical removal after serving their purpose. I will first discuss the mechanical and electrochemical properties of MgZnCa glasses, and illustrate their implant degradation performance. Because of the improved alloying possibilities for glasses (in contrast to crystalline alloys), their Zn-content can be varied widely, and alloys with more than 28 at.% Zn show strongly reduced degradation in simulated body fluid owing to passivating Zn-oxide layer formation [1]. To study tissue response to the degrading alloys we performed in-vivo experiments by implanting glassy [2] and fine-grained crystalline [3, 4] MgZnCa pins in rat femurs. The degradation and bone response were followed in situ by microcomputed x-ray computer tomography (mu;CT), which has the advantage of being non-invasive. The mu;CT studies show that bone recovers completely after implant degradation, even in cases of previous massive bone alteration. These studies also show that Mg generates enhanced bone neoformation, giving evidence of its good osteoconductivity and osteoinductivity. Based on these results, I will discuss the potential of Mg-based BMGs for biodegradable implant applications.
[1] B. Zberg, P. J. Uggowitzer, J. F. Löffler, Nature Mater. 8 (2009) 887.
[2] V. Wessels et al., Adv. Eng. Mater. 14 (2012) B357.
[3] T. Kraus et al., Acta Biomater. 8 (2012) 1230.
[4] S. F. Fischerauer et al., Acta Biomater. 9 (2013) 5411.

Based on a correlation between the local structural motifs and the propensity for shear transformations, Zr-rich compositions in Zr(Ti)-Cu-Al bulk metallic glasses are expected to be more prone to spread-out plastic deformation and more profuse shear banding. This structural perspective has guided us to land a Zr-Ti-Cu-Al BMG that exhibits record-breaking fracture toughness, on par with that of the Pd-based BMG recently developed at Caltech. For BMGs, damage tolerance is a key mechanical property needed for their practical use. To reach a high fracture toughness, the only route reported so far is to compositionally base the BMG on the precious palladium (Pd) that has an exceptionally high Poisson&’s ratio (0.42). Our new Zr61Ti2Cu25Al12 (ZT1) BMG is as tough, but at the same time consists of common engineering metals and has robust glass-forming ability (cm in size). The new BMG, while having an unimpressive Poisson&’s ratio of 0.367, derives its high toughness from its high propensity for crack deflection and local loading-mode change (mixed mode I and mode II) at the crack tip due to extensive shear band interactions. The crack-resistance curve (R-curve) of this BMG has been obtained following ASTM standards, employing both “single-specimen” and “multiple-specimen” techniques as well as fatigue pre-cracked samples.

Indented metallic glasses in the nano-scale deform via shear strain bursts with a broad range of sizes. Conventional continuum descriptions are not expected to apply to such highly stochastic, intermittent deformations. Here, we show that a statistical analysis of the strain bursts in five metallic glasses with different hardness (strength) values. A concordant region model is introduced to establish the dependence of the cut-off of strain burst size on the deformation units. For the soft metallic glasses with smaller deformation units, the cut-off of the strain burst size truncates the scale-free behavior at larger strain burst sizes. For the hard metallic glasses, the scale-free behavior occurs in a wide range of strain burst size, indicating a self-organized critical dynamic behavior.

It has been widely acknowledged that metallic glass composites are very complex,
and a broad range of contributions have been suggested to explain the toughening
mechanism [1,2]. This includes the shear modulus, morphology, size, spacing, volume
fraction of the second phase, and strength and toughness of the interface. Previous
studies suggest these contributions, however, do not provide quantitative experimental
evidence.
Within this approach, we paid tribute to the complexity of the toughening mechanism by revealing quantitatively the correlation between plastic zone size (RP) and second phase spacing (s) [3]. Furthermore, we emphasized the importance of morphology, and identify and quantify the ideal heterostructure, which maximizes strength and toughness. The fracture strength of the optimized MG heterostructure almost reached to the fracture strength of the monolithic MG while fracture strain and overall toughness of the MG heterostructure became 3.75 and 5 times higher than that of the monolithic MG, respectively. The second phase elasticity and shear modulus were also found to be contributing to the overall elasticity. We identified the pores&’ ratio of diameter to spacing (d/s) as one of the major factors controlling the mechanical properties of MG heterostructures, which is most efficient when d/s asymp; 1. Effectiveness of MG heterostructures were also determined to be depending on the size of the sample, w, in comparison to s. Our experimental findings illuminate the complexity in MG composites, which can be resolved with our artificial microstructure approach, and thereby, further guide to improve existing heterostructures and design novel structures with predictable properties.
1. Sarac, B., Ketkaew, J., Popnoe, D.O. & Schroers, J. Honeycomb Structures of Bulk Metallic Glasses. Adv. Funct. Mater. 22, 3161-3169 (2012).
2. Sarac, B. & Schroers, J. From brittle to ductile: Density optimization of Zr-based bulk metallic glass cellular structures Scr Mater 68, 921-924 (June 2013).
3. Sarac, B. & Schroers, J. Designing Tensile Ductility in Metallic Glasses. Nat. Commun. (2013 (In Press)).

Metallic glasses are commonly known to undergo brittle failure at room temperature, where deformation is confined to very narrow shear bands. However, in our previous work [J. Appl. Phys. 103 (2008) 083539], we could demonstrate an apparent size-dependent brittle-to-ductile transition. Subsequently, this has been confirmed or dismissed by several research groups and remains a controversial topic.
In this work we continue our search for origins on the apparent change in deformation behavior, and present further results from nanomechanical testing performed on more than 200 Pd₇₇Si₂₃ metallic glass specimen of sizes between 200 and 2000 nm. We find that in pillars with diameters of ~400 nm and smaller, a homogeneously deformed region evolves at the pillar&’s top end, while concomitant shear banding seems to be more and more suppressed. A careful analysis of SEM micrographs of pillars above this critical size reveals that shear-band spacing scales with sample size. Assuming a shear-band thickness of some few tens of nm, this indicates that the shear-band volume fraction approaches the order of unity at the apparent brittle-to-ductile transition. In addition, we performed rate dependent microcompression tests varying the applied rate over five orders of magnitude. Post-mortem SEM micrographs indicate the emergence of homogeneous deformation at lower rates and intense shear banding at the highest rates for all sizes. In fact, at the slowest rates, an apparent brittle-to-ductile transition is observed at sample-sizes as large as 1000 nm. We find that the strain rate exponent m is independent of sample size and thus conclude, that the emergence of an apparently homogeneous deformation mode does not indicate a change in deformation mechanism. Instead, our results indicate that the apparent transition in flow behavior can be explained by a coalescence of shear bands due to a size-dependent reduction of the shear band spacing that approaches the shear band dimensions itself.

At low homologous temperatures bulk metallic glasses (BMGs) are known to deform inhomogeneously via the formation and operation of shear bands (SBs). The operation of these bands is intermittent, as reflected in serrated flow curves, and can be understood in the context of a stick-slip process. Below a critical temperature, however, a transition from serrated to non-serrated flow behavior is observed [1]. In this talk I will present experiments on shear banding in BMGs where we combine time-resolved compression tests with in-situ acoustic emission (AE) monitoring. By measuring the time scales of the serrations as a function of temperature we find that stable shear-banding in BMGs is thermally activated and has velocities in the mm/s range at room temperature [2]. Further, we find that one AE signal precedes each load drop in the serrated flow curve, which shows that AE reveals the moment of SB initiation. By inferring a model picture from stick-slip in granular media we can attribute this SB initiation to a mechanism of structural dilatation of a few per cent [3], which is the typical excess free volume found in the undercooled liquid region near the glass transition. Finally, by performing slide-hold-slide compression experiments, we are able to characterize the static (re-)strengthening of a SB [4]. We can describe this via an effective temperature model and obtain an activation energy of 0.16 eV for such SB healing. Using this model, we can also explain the above-mentioned transition from serrated to non-serrated flow below a critical temperature.
[1] A. Dubach, F. H. Dalla Torre, J. F. Löffler, Acta Mater. 57 (2009) 881.
[2] R. Maass, D. Klaumünzer, J. F. Löffler, Acta Mater. 59 (2011) 3205.
[3] D. Klaumünzer et al., Phys. Rev. Lett. 107 (2011) 185502.
[4] R. Maass et al., Appl. Phys. Lett. 100 (2012) 071904.

Under certain temperatures or strain rates, bulk metallic glasses (BMGs) could exhibit flow serrations in the stress-strain curves of compression experiments. The transition between the serrated flow and non-serrated flow implies that different mechanisms dominate during their inhomogeneous deformation. Furthermore, the statistical information hidden behind the "disordered" serration can be used as an indicator for the microstructure evolution. In this study, we will review recent work on the serration behavior of BMGs, and summarize the statistical characteristics of serrated flows. The cases with different loading conditions will be discussed. Based on the mean-field theory, a simple slip-avalanche model is being developed to describe and predict the serration behavior of BMGs under various strain rates.

Progress in micro- and nanofabrication techniques enables the creation of architectured microlattices with dimensional control over six orders of magnitude, from centimeters down to nanometers. This hierarchical control allows exploiting size effects in mechanical properties of materials to create better structural materials. We present the fabrication, characterization and properties of hollow metallic glass nickel-phosphorous microlattices. Their wall thicknesses, deposited by electroless plating, range from ~ 60 nm up to 600 nm, resulting in relative densities spanning 0.02% to 0.2%. Uniaxial quasi-static compression tests revealed two different regimes in maximum stress: (1) Structures with a wall thickness above 150 nm failed by catastrophic failure at the nodes and fracture events at the struts, with significant micro-cracking; (2) Lattices whose wall thickness was below 150 nm failed initially via buckling followed by significant plastic deformation rather than by post-elastic catastrophic fracture. This departure in deformation mechanism from brittle to deformable exhibited by the thin-walled structures is discussed in the framework of brittle-to-ductile transition emergent in nano-sized metallic glasses.

We present a novel artificial microstructure approach, which allows us to fabricate a broad range of cellular architectures, from perfectly periodic to highly stochastic. For these cellular structures, we introduce imperfections that we can precisely control and evaluate the effect on their mechanical response. Altogether, we fabricated approximately 150 different cellular structures. The mechanical performance of the periodic structures is generally superior over the stochastic structures. In contrast, the tolerance to structural defects of the stochastic structures is much higher than the periodic structure. We revealed that such different behavior originates from the different overall deformation modes of periodic and stochastic structures. Within the stochastic structures, more diffusive deformation patterns, more uniform plastic strain distribution, as well as high “stress distribution stability” are speculated to contribute to the distinguished flaw tolerance in comparison to the periodic structures. Within the periodic structures, plastic deformation is collectively localized in narrow “deformation bands”. Furthermore, as an example of a stochastic structure, we considered a bio-cuticle structure. Characterization of this structure suggested that nature also considers flaw tolerance in its selection process. Our approach and findings help in the design and selection process of cellular materials, particularly when flaw tolerance is considered.

High strength combined with low stiffness and high hardness are appealing properties making bulk metallic glasses (BMGs) candidate materials for many structural applications. However, as strength and toughness are generally mutually exclusive, early glasses presented a challenge with brittle fracture behavior often resulting from single shear band failure and toughness values in some cases as low as 3 MPa.m1/2. BMG-matrix composites show a significantly improved fracture behavior as crystalline second-phase dendrites can arrest shear bands and stop the formation of cracks before they exceed critical size. Indeed, in 1.2-1.5 GPa strength Zr-Ti-Nb-Cu-Be glass-composites, failure occurs after subcritical crack growth in a highly ductile manner showing marked crack-resistance curve (R-curve) behavior with toughnesses as high as 150 MPa.m1/2. In monolithic BMGs, the production of glasses with high bulk-to-shear modulus ratios, as shown by the example of the 1.5 GPa strength Pd79Ag3.5P6Si9.5Ge2 monolithic glass, can lead to multiple shear band formation and inhibited cavitation to form cracks; the resulting shielding again causes extensive R-curve behavior and toughness values up to 200 MPa.m1/2. However, while the toughness of these materials were determined solely in three-point bending, the fracture toughness of the semi-brittle Zr52.5Cu17.9Ni14.6Al10Ti5 monolithic glass was found to vary considerably (from ~25 to 95 MPa.m1/2) when tested with different specimen geometries, which appeared to change the extent of multiple shear band formation at the crack tip. Here, we report on a systematic study to compare the fracture toughness of three different glasses - one low toughness (< 50 MPa.m1/2), one medium and one high toughness (>150 MPa.m1/2) - which were tested under different loading conditions, specifically deep-cracked bending vs. edge-cracked tension, in order to understand the influence of such variables as sample size, geometry and stress-state on the measured fracture toughness of BMGs, as well as to determine the origin of the large variations in toughness often found for these materials.

Fracture toughness is one of the crucial material parameters with respect to the use of bulk metallic glasses (BMGs) as structural materials. From an application-oriented view, only a few BMGs feature sufficiently high fracture toughness values. However, most of these investigated alloys are not tested under standard ASTM conditions. This is due to limited sample thickness and the measurement of notch toughness without proper fatigue pre-cracking. Most recently, feasible R-curves of ductile Pd- and Zr-based BMGs have been measured yielding promising values up to 200MParadic;m. In this study, the R-curves of Pd- and Zr-based BMGs with high base metal content are presented. Thereby, special consideration is also given to various relevant experimental parameters such as the controlled introduction of a straight fatigue pre-crack.

Materials with both high strength and high toughness are required for applications in lightweight and high-performance engineering structures. The strength of bulk metallic glasses (BMGs) are higher than any reported bulk alloys. However, low damage tolerance in some systems limit their use in engineering applications. Extensive work has been conducted to search for tough monolithic BMGs and dendrite-reinforced BMG matrix composites. The present work extends beyond composition design and illustrates that the toughness of both tough and brittle BMGs can be improved by introducing ‘designed heterogeneity&’ that is effective in deflecting crack growth from a nominal fracture plane. It is shown that the initiation of fracture instability in a brittle BMG has been changed, resulting in a tougher BMG and producing a large dimple size on the fracture surface, consistent with the observation of increased toughness.

Bulk metallic glasses (BMG) can exhibit favorable properties relative to traditional crystalline metals such as near theoretical strength coupled with metal like fracture toughness values. However, BMGs also exhibit unique characteristics such as 1) fatigue endurance strengths that can range widely for different compositions with similar tensile strength and 2) ductility and strain softening/hardening behavior that is dependent on both sample size and loading configuration. In order to expand the application of BMGs into more structural and mechanical components designers need to ensure the components will not fail unexpectedly. To achieve this a detailed understanding of the unique fracture and fatigue characteristics of BMGs is needed. The BMG with composition Zr_52.5Cu_17.9Ni_14.6Al₁₀Ti₅ exhibits excellent fatigue endurance strength in four point bending with a ratio of 10⁷ cycle endurance strength to ultimate tensile strength of ~0.25, although small amounts of casting porosity can reduce the endurance limit by ~60%. This behavior was attributed to a relatively high fatigue threshold (ΔK_TH asymp; 2 MParadic;m) and an insensitivity to degrading environmental effects in ambient air compared to other Zr-based BMGs that exhibit very poor fatigue endurance strengths. Also, other factors found to influence the fatigue response include structural relaxation and residual stresses. Next, the issue of size dependent fracture behavior was also examined for the Zr_52.5Cu_17.9Ni_14.6Al₁₀Ti₅ BMG and it was found that when sample size requirements of ASTM standard E399 were met the fracture toughness values are sample size and geometry independent. However, if ASTM standard E1820 is applied to allow smaller sample dimensions the results become sample size dependent in a manner that is distinct from crystalline metals. The origins and implications of this size dependence will be discussed.

Widespread adoption of metallic glasses in applications motivated by high strength and elasticity combined with plastic-like processing has been stymied by their lack of tensile ductility. One emerging strategy to couple the attractive properties of metallic glasses with resistance to failure by shear localization is to employ sub-micron sample or feature length scales, although conflicting results shroud an atomistic understanding of the responsible mechanisms in uncertainty. The inherently far-from-equilibrium nature of metallic glasses additionally provides various structural states for the material to occupy, from structurally relaxed to rejuvenated. Thus, within a given alloy, the mechanical and physical properties strongly depend on extrinsic factors such as thermal and mechanical history.
We report in situ deformation experiments of directly molded Pt57.5Cu14.7Ni5.3P22.5 metallic glass nanowires, which show tunable tensile ductility. Initially brittle as-molded nanowires can be coerced to a distinct glassy state upon irradiation with Ga+ ions, leading to tensile ductility and quasi-homogeneous plastic flow. This behavior is reversible and the glass returns to a brittle state upon subsequent annealing. By systematically varying the irradiated volume fraction and ion fluence, we identified both a transition in plastic deformation behavior and reductions in the measured yield strength relative to the as-molded state. Whereas signatures of changes to the amorphous structure owing to ion irradiation as obtained from electron diffraction are undetectable, the yield strength reduces by 100s of MPa. To reconcile changes in mechanical behavior to glass properties, we adapt models equating the released strain energy during shear banding to a transit through the glass transition temperature and incorporate the excess enthalpy associated with distinct structural states. Our model suggests that ion irradiation increases the fictive temperature of our glass by 10s of degrees; the equivalent of several orders of magnitude change in cooling rate. Our results suggest a novel mechanism for homogenous plastic flow in nano-scaled metallic glasses and new processing strategies leading to unprecedented control of mechanical properties.

Monolith metallic glasses are usually brittle under tension, severely limiting their practical applications. How to impart ductility to these brittle materials is a tremendous challenge. Via nanomechanical tests of model metallic glasses using molecular dynamics simulations, we show one can enhance intrinsic ductility of glasses by controlling the bonding and packing, or produce extrinsic ductility of glasses by controlling the sample size and surfaces. These results are also common among other amorphous solids such as amorphous silicon and silica. The importance of sample processing in numerical and experimental nanomechanical tests will be highlighted.

Metallic glasses have been tested at different temperatures, strain rates, and under different stress states in order to determine the effects of changes in such variables on the strength and/or plasticity. Stress state changes were accomplished via testing smooth and/or notched samples in tension/compression with superimposed pressure over a range of test temperatures. While minimal effects of superimposed pressure on the strength/plasticity was obtained at temperatures well below the glass transition, Tg, significant effects of pressure + superimposed pressure were exhibited near Tg. Both micro- and nano- metallic glass samples were prepared and tested in tension to determine the effects of changes in sample size on plasticity. The effects of such changes on the plasticity will also be described.

We demonstrate the fabrication of 100-nm diameter and 650-nm gauge length nano-tensile samples of Ni-P metallic glasses via template electroplating. Samples with equivalent dimensions and geometries were also carved out from an identically-electroplated, 650nm-thick Ni-P film via Focused Ion Beam (FIB). XRD, EDS, and TEM analysis revealed that both types of samples were amorphous, with nearly identical chemical composition. Room-temperature in-situ uniaxial tensile experiments revealed the emergence of plastic deformability in 100nm-diameter samples fabricated by both techniques. Necking was found to always precede final failure, which always commenced via shear banding. FIB-miled samples displayed nearly a factor-of-three higher post-elastic deformability before failure. In contrast, 500nm-diameter samples fabricated by both techniques failed catastrophically, similar to bulk. Molecular dynamics simulations performed on similar unirradiated and “irradiated” Ni80Al20 systems corroborate experimental results. These findings suggest that the emergence of brittle-to-ductile behavior in nano-sized Ni-based metallic glasses is caused by size effect while the amount of ductility is determined by surface energy state.