Bulk Metallic Glasses

Bulk metallic glasses (BMGs) in metal-metal systems such as La^-, Mg^-, and Zr^-based alloys were first prepared in the early 1990s by the stabilization of supercooled liquid. Since then much effort has been devoted to the development of BMGs for both fundamental scientific research and for industrial applications. As a result, many unique and useful properties of BMGs have been found. In particular, research at the Institute for Materials Research has been concentrated primarily on early transition metal (Zr^-, Ti^-, and Hf^-based) systems, lanthanide metal (Ln^-based) systems, simple metal (Mg^- and Ca^-based) systems, and noble metal (Pd^- and Pt^-based) systems. Because of their excellent properties, BMGs are expected to emerge as a new type of industrial or engineering material. The development of late transition metal (LTM)-based BMGs is strongly encouraged due to material costs and the availability of raw material deposits. Therefore, an Fe-based BMG in the Fe–Al–Ga–P–C–B alloy system was successfully developed in 1995. Also at that time, three empirical component rules for the stabilization of a supercooled metallic liquid were proposed. These rules stated that (1) the multicomponent system should consist of three or more elements, (2) there should be a significant difference (greater than ~12%) in the atomic sizes of the main constituent elements, and (3) the elements should have negative heats of mixing. A variety of Fe-based, Co-based, Nibased, and Cu-based BMGs have been synthesized in accordance with these rules and other topological and chemical criteria. As a result, various unique properties of LTM-based BMGs have been obtained. These properties have not been obtained in any crystalline alloys. Therefore, it should be possible to extend the range of applications. This chapter reviews recent results on the formation, properties, thermal stability, workability, and applications of LTM-based BMGs.

When we try to find out the structure–property relationships for metallic glasses, we feel lost and have trouble knowing where to begin, because describing the atomic structure of a glass is already a major challenge. The structure of glass is called amorphous, meaning shapeless, a term that refuses rigorous characterization. Actually it is a great accidental gift of nature that many substances are crystalline, so that we can discuss their structure in such a simple way, such as the lattice, symmetry, and the unit cell, even though there are so many, of the order of 10 atoms cm^-3 in a crystal. In this chapter, we start with the most basic question of how to describe the structure of liquids and glasses, and discuss how we can start constructing a theory that can describe the structure–property relationships of metallic glasses. We will take a local, rather than global, view of the structure, and consider how the local structure is related to the local properties.

A major advantage of atomistic simulations is that a detailed picture of the model under investigation is available, and so they have been very instrumental in explaining the connection of macroscopic properties to the atomic scale. Simulations play a significant role in the development and testing of theories. For example, simulations have been extensively used to test the mode-coupling theory (MCT). The theory predicts that at some critical temperature T _c, known as the mode-coupling temperature, the supercooled liquid undergoes a structural arrest, prohibiting the system from accessing all possible states, thus, essentially undergoing an ergodic to nonergodic transition. It gives definite predictions on various correlation functions that can be calculated directly in simulations. Simulations and MCT have played a tremendous role in elucidating a majority of what we now understand about the dynamics of glass-forming systems.

Simulations can also be used to compare with experimental results to validate the model, so that one can use simulation results to measure properties not accessible to experiments. In many cases, as will be illustrated in the next sections, results of simulations motivate experimental investigations. Part of the goal of this chapter is to examine the contributions of atomic simulations to the current state of understanding of metallic glasses.

The emergence of synthetic bulk metallic glasses (BMGs) as a prominent class of functional and structural materials with a unique combination of properties has been an important part of the materials science scene over the past two decades. To date, a number of BMGs have been successfully developed and commercialized for engineering applications utilizing their exceptional properties. However, one of the biggest stumbling blocks for the use of these noncrystalline alloys is still the low glass-forming ability (GFA) of many systems, which is a long-standing problem that is far from being adequately solved. Understanding the nature of glass formation and GFA is the key to developing new BMGs with improved properties and economic manufacturability for industrial applications.

Appreciable progress has been made in understanding the physical insights of bulk glass formation, such as the crystallization mechanisms, alloying effects, liquid fragility, etc., and the macro- and microdeformation mechanisms of BMGs. In this chapter, the focus is placed on how to effectively quantify and represent relative GFA of different glass-forming systems. Previous work on the known GFA indicators and a comprehensive review of recent developments in this area are summarized. One of the main emphases is the establishment of the γ parameter and the demonstration of its better reliability and applicability over all previous GFA indicators. In particular, underlying mechanisms and physical insights of the effective γ criterion will be analyzed in detail. Future directions in understanding and measuring GFA of metallic alloys will also be surmised. Specifically, this chapter contains the following sections:

Microstructural characterizations are performed to investigate both the atomic structure and the stability of bulk metallic glasses (BMGs). Many characterizations with a variety of instruments are performed to identify the phases produced during preparation (i.e., quenching), annealing, and devitrification of BMG. The typical parameters that are quantified are the size, morphology, composition, crystal structure, and volume fraction of the phases formed. These parameters may be used to improve the alloy design process. Some techniques provide continuous monitoring of time-dependent processes, such as crystallization, as a function of time and temperature. High-resolution techniques are used to investigate the atomic structure of the glass including short- and medium-range order. Specialized characterizations may be performed to investigate the quantity and type of free-volume in the glass. Microstructural characterizations may also be performed in conjunction with mechanical property tests to investigate the type of failure, the source of crack initiation, and the interaction of shear bands that are produced under stress with microstructural features.

In this chapter, many of the techniques that have been used to characterize the microstructures of BMGs are reviewed. The techniques are divided broadly into the routine techniques, such as differential scanning calorimetry, X-ray diffraction, and scanning electron microscopy, that are used to evaluate parameters such as the glass-forming ability, the critical dimension, and examine the general microstructure, and the higher resolution, more specialized techniques, such as high-resolution and fluctuation electron microscopy, field ion microscopy, atom probe tomography, small angle scattering, and positron annihilation spectroscopy, that are used to characterize the nanometerscale structure, phase composition, open volume, etc.

Although metallic glasses synthesized by rapid quenching from the melt were first discovered in 1960 by Duwez and coworkers at Caltech, the study of the mechanical behavior of metallic glasses only started in the early 1970s. Metallic glasses were found to deform elastically and exhibit negligible plasticity in uniaxial tension. Despite a limited macroscopic tensile plastic strain (<0.5%), exceptionally high strain (~100) was observed to take place within localized shear bands. One of the scientific questions naturally arises: how does a shear band nucleate and propagate in a medium presumably consisting of randomly packed atoms? Several theories, including the freevolume model and the dislocation model, were subsequently proposed to address the shear band formation and propagation. It was impossible to carry out irrevocable experiments to prove these theories at that time due to the limited size of the samples, and thus a poorly defined stress state during mechanical testing, and the lack of advanced analytical tools. However, the situation changed after the successful development of bulk metallic glasses (BMGs) in many alloy systems, e.g., Zr^-, Mg^-, Pd^-, La^-, Cu^-, Ti^-, and Febased. As a result of this development, research of BMGs has been very active, especially in the area of mechanical deformation.

Structural components are frequently subjected to repeated or cyclic loading. The resulting cyclic stresses, which may be far below the ultimate tensile strength of materials, can result in a microscopic physical damage to the material. The microscopic damage can accumulate with continued cyclic loading until it develops into a crack that could lead to the catastrophic failure. This process of damage and failure due to cyclic loading is called fatigue.

Fatigue is a dynamic phenomenon that accounts for ~90% of all service failures from mechanical causes. In general, the fatigue life involves the number of loading cycles to initiate and propagate a crack to a critical size. Fatigue failure occurs in three stages: crack initiation, stable crack growth, and fast fracture. The main factors that contribute to fatigue failures include the number of load cycles, stress range, mean stress, and local stress concentrations. It is necessary to take these factors into account in the design of materials for structural components.

There has been a growing interest in the corrosion behavior of bulk metallic glasses (BMGs). The enhanced glass formability of these amorphous alloys has made the fabrication of bulk size components easier and thereby improving their prospects for engineering applications. The corrosion behavior of BMGs is particularly pertinent when considering biomedical applications and is also relevant for other applications, e.g., watches and electronic casings. The corrosion resistance is also a critical factor for the consideration of their use in hostile or chemical environments.

The corrosion properties of an amorphous alloy are expected to be superior to those of its crystalline counterpart due to its chemical homogeneity and lack of microstructure. Amorphous alloys lack grain boundaries, dislocations, and other defects that are commonly the culprits behind the localized corrosion observed in crystalline alloys. The rapid cooling rates required to produce amorphous alloys are believed to promote chemical homogeneity since there is no time for appreciable solid-state diffusion. The short time available for significant diffusion suggests that amorphous alloys should lack second phases, precipitates, and segregates. However, this assertion has been shown not to always be true for BMGs. It should be noted that the second phases (crystalline inclusions) observed in some BMGs are a result of heterogeneous nucleation often caused by impurities in the melt.