Using state-of-the-art lab techniques and powerful
computer simulations, Johns Hopkins researchers have
discovered how atoms pack themselves in unusual materials
known as metallic glasses. Their findings should help
scientists better understand the atomic scale structure of
this material, which is used to make sports equipment, cell
phone cases, armor-piercing projectiles and other
The discovery, marking the culmination of a two-year
research project, was reported in the Jan. 26 issue of the
journal Nature. The work represents a major step
forward because the tools used to study traditional
crystalline metals do not work well with metallic glass,
and a better understanding of the material has been sorely
"How the atoms pack themselves in metallic glass has
been a mystery," said Howard Sheng, an associate research
scientist in the Whiting School's
Materials Science and Engineering and lead author of
the Nature paper. "We set out to decipher this packing
information, and we were ultimately able to provide a clear
description of how the atoms arrange themselves in metallic
In conventional metals, atoms crystallize into uniform
three-dimensional patterns known as lattices. But about a
half-century ago, materials scientists learned how to make
glassy metals by cooling a metallic liquid so quickly that
the internal atomic configurations froze before the atoms
had a chance to arrange themselves into a lattice pattern.
The new material was described as amorphous, meaning its
atoms seemed to be arranged in an irregular fashion without
the long-range order characteristic of crystalline
materials. This amorphous atomic structure is commonly
found in other materials such as window glass, but it
rarely occurs in metals.
Unlike windowpanes, metallic glasses are not
transparent or easy to shatter. Many traditional metals are
easy to bend out of shape because of defects (dislocations)
in their crystal lattice. But metallic glasses have no
crystal lattice and no such dislocations, and their
disorderly arrangement of atoms gives them distinctive
mechanical and magnetic properties. Metallic glasses, which
are usually made of two or more metals, can display great
strength, large elastic strain and toughness. Another
advantage is that, like weaker plastic materials, they can
easily be heated, softened and molded into complex
For their experiments, the
researchers made small samples of various metallic glasses.
The sample pictured here is a nickel-phosphorous metallic
PHOTO BY HIPS/WILL KIRK
Despite the great potential of metallic glasses, the
researchers who make them have been hampered by a scarcity
of basic science knowledge about the materials. Powerful
transmission electron microscopes can be used to view rows
of atoms lined up in traditional metals. But when these
instruments are used on a metallic glass, the resulting
image is one of a scattered array of atoms, forming no
obvious pattern. Because so little has been known about how
atoms are arranged in metallic glasses, a number of basic
materials science problems, such as how a metallic glass
deforms, remain unsolved.
To help fill the knowledge gap, a team supervised by
Evan Ma, a professor of materials science and engineering,
launched a two-pronged approach to solve the mystery of how
metallic glass atoms are arranged. "Our goal was to advance
the understanding of atomic packing in metallic glasses,"
Ma said. "This is a difficult task because of the lack of
long-range order in these amorphous structures. Yet it is
of fundamental importance because it is the structure that
The researchers made samples of a number of binary
metallic glasses, each composed of two elements, and then
subjected them to high-tech lab tests to gather information
about the samples' three-dimensional atomic configurations.
Some of these experiments, conducted at Oak Ridge and
Brookhaven national laboratories, involved X-ray
diffraction and extended X-ray absorption fine structure
data taken at synchrotron X-ray sources. Other analyses,
utilizing a method called reverse Monte Carlo simulations,
were conducted with a computer cluster at Johns Hopkins.
Independent of these lab tests, the researchers used
powerful computer resources provided by the National Energy
Research Scientific Computing Center to run virtual
experiments aimed at uncovering the arrangement of metallic
glass atoms. Results from the lab experiments and the
computer trials were used to validate one another,
confirming the researchers' conclusions.
One of their key findings was that metallic glass
atoms do not arrange themselves in a completely random way.
Instead, groups of seven to 15 atoms tend to arrange
themselves around a central atom, forming three-dimensional
shapes called Kasper polyhedra. Similar shapes are found in
crystalline metals, but in metallic glass, the researchers
said, these polyhedra are distorted and do not align
themselves in long rows. In metallic glass, the polyhedra
join together in unique ways as small nanometer-scale
clusters. In the journal article, these structural features
were described as chemical and topological short-range
order and medium-range order.
The Johns Hopkins engineers also made important
discoveries about how low-density spaces form among these
clusters in metallic glass. These "cavities" affect the way
the material forms as a glass and the mechanical properties
it will possess.
Sheng, the lead author of the journal article,
believes these discoveries will lead to significant
advances in the understanding of metallic glass. "Our
findings," he said, "should allow the people who make
metallic glass to move closer to intelligent design
techniques, developing materials with the precise
mechanical characteristics needed for specific products.
The discoveries also advance our understanding of materials
science in general."
The metallic glass research was funded by a grant from
the U.S. Department of Energy. Along with Sheng and Ma, the
authors of the Nature article are Weikun Luo, a
doctoral student in the Department of Materials Science and
Engineering at Johns Hopkins; F.M. Alamgir, of the National
Institute of Standards and Technology; and J. M. Bai, of
the Oak Ridge National Laboratory.