Research Project

Metallic Glasses

Research in our group has covered a wide range of topics in metallic glasses, ranging from alloy design (including the development of novel metallic-glass-matrix composite materials) and studies of the atomic-scale structure to very practical studies of mechanical behavior, including both deformation and fracture.

Metallic glasses or amorphous metals are novel engineering alloys in which the structure is not crystalline (as it is in most metals) but rather is disordered, with the atoms occupying more-or-less random positions in the structure. In this sense, metallic glasses are similar to the more familiar oxide glasses such as the soda-lime glasses used for windows and bottles.


From a practical point of view, the amorphous structure of metallic glasses gives them two important properties. First, like other kinds of glasses they experience a glass transition into a supercooled liquid state upon heating. In this state the viscosity of the glass can be controlled over a wide range, creating the possibility for great flexibility in shaping the glass. For example, Liquidmetal Technologies  produced the golfing putter shown here:



Second, the amorphous atomic structure means that metallic glasses do not have the crystalline defects called dislocations that govern many of the mechanical properties of more common alloys. The most obvious consequence of this is that metallic glasses can be much stronger (3-4 times or more) than their crystalline counterparts. Another is that metallic glasses are somewhat less stiff than crystalline alloys. The combination of high strength and low stiffness gives metallic glass very high resilience, which is the ability to store elastic strain energy and release it. This is dramatically illustrated in this video:



In the video, identical ball bearings are dropped onto a metallic glass (left) and a piece of stainless steel (right). The high strength and low stiffness of the glass alloys the ball bearing to bounce for a long time, while in the stainless steel the low strength causes plastic deformation which quickly damps out the kinetic energy of the ball bearing.


From a scientific point of view, metallic glasses are fascinating because many of their important properties and behavior are only now beginning to be understood. Part of the challenge in understanding them comes about because it is much more difficult to characterize the structure (and, critically, the defects in the structure) of an amorphous material than it is of a crystalline material. Research in our group has covered a wide range of topics in metallic glasses, ranging from alloy design (including the development of novel metallic-glass-matrix composite materials) and studies of the atomic-scale structure to very practical studies of mechanical behavior, including both deformation and fracture.


In some of our recent work, we are trying to understand the basic mechanisms by which metallic glasses fracture, which of obvious importance for engineering applications. Metallic glasses are unlike crystalline alloys in that even though they have little or no ductility they can still be relatively tough (resistant to fracture). We conducted a series of experiments in which we performed fracture studies of metallic glass specimens at a synchrotron, which produce extremely intense beams of x-rays that we use to probe how the structure of the glass evolves as we increase the stress. The basic idea is shown here:



The metallic glass sample has a notch with a crack at its tip. Loading the sample at the three points shown causes bending which increases the stress at the crack tip. We using the x-ray beam to produce diffraction rings, from the shape of which we can deduce the strain (or stress) on the sample at any given point, and by moving the specimen we can map out these strains as a function of position around the crack tip. Analysis of these data suggests that the glass can develop a substantial region of plastic deformation around the crack tip at room temperature, but not when it is cooled to cryogenic temperatures.


Metallic glasses are different from ordinary (crystalline) alloys in that at high stresses deformation tends to be concentrated into narrow regions called “shear bands,” like those shown here:



Amazingly enough, shear band dynamics have a lot in common with deformation with a much broader range of materials, from metal single crystals to granular materials (such as sand) and the earth (in earthquakes):


credit J.T. Uhl (Scientific Reports 5, 16493 (2015))


In each of these cases, deformation occurs through individual slip events which, collectively, can result in a “slip avalanche.” In collaboration with Wendy Wright (Bucknell University), Karin Dahmen (University of Illinois), and Dmitry Denisov and Peter Schall (University of Amsterdam), we have shown that the dynamics of these slip avalanches obey the same scaling behavior, despite the vast differences in length- and time-scales over which they occur. For example, here’s a plot of the distribution of avalanche sizes for metallic glasses (BMG) compared to other materials:


credit J.T. Uhl (Scientific Reports 5, 16493 (2015))


A simple model (developed by Karin Dahmen) explains this behavior in terms of slipping “weak spots” in the material, independent of the actual mechanism underlying the effect.  The agreement between the model and experiments suggests universal aspects of behavior in these systems which we are just beginning to understand. This basic result may extend to deformation of other materials (such as porous materials and foams). It may allow us to predict slip behavior at high stresses by extrapolating from data collected at lower stresses and provide a quantitative basis for further applications, such as hazard prediction studies of earthquakes on much larger tectonic scales.


This research is funded by the National Science Foundation (NSF). Earlier work on metallic glasses in our group was supported by the Department of Energy (DOE), the Army Research Office (ARO), and the Army Research Laboratory (ARL).



Additional reading:

  • V. Denisov, Scientific Reports 7, 43376 (2017)

doi: 10.1038/srep43376

  • J. Wright et al. J. Appl. Phys. 119, 084908 (2016)

doi: 10.1063/1.4942004

  • C. Hufnagel et al., Acta Mater. 109, 375 (2016)

doi: 10.1016/j.actamat.2016.01.049

  • T. Uhl et al. Scientific Reports 5, 16493 (2015)

doi: 10.1038/srep16493

  • Antonaglia et al., Phys. Rev. Lett. 112, 155501 (2014)

doi: 10.1103/PhysRevLett.112.155501

  • K. Slaughter et al., APL Materials 2, 096110 (2014)

doi: 10.1063/1.4895605

  • C. Hufnagel et al., PLOS ONE 8, e83289 (2013)

doi: 10.1371/journal.pone.0083289


Back to top