Strong, tough and now cheap: New way to process metallic glass developed (w/ video)

May 12, 2011 by Marcus Woo
The metallic-glass rod before heating and shaping (left); the molded part (middle); the final part trimmed of excess material (right). Credit: Marios D. Demetriou

(PhysOrg.com) -- Stronger than steel or titanium -- and just as tough -- metallic glass is an ideal material for everything from cell-phone cases to aircraft parts. Now, researchers at the California Institute of Technology have developed a new technique that allows them to make metallic-glass parts utilizing the same inexpensive processes used to produce plastic parts. With this new method, they can heat a piece of metallic glass at a rate of a million degrees per second and then mold it into any shape in just a few milliseconds.

"We've redefined how you process metals," says William Johnson, the Ruben F. and Donna Mettler Professor of Engineering and Applied Science. "This is a paradigm shift in ." Johnson leads a team of researchers who are publishing their findings in the May 13 issue of the journal Science.

"We've taken the economics of plastic and applied it to a with superior engineering properties," he says. "We end up with inexpensive, high-performance, precision net-shape parts made in the same way plastic parts are made—but made of a metal that's 20 times stronger and stiffer than plastic." A net-shape part is a part that has acquired its final shape.

Metallic glasses, which were first discovered at Caltech in 1960 and later produced in bulk form by Johnson's group in the early 1990s, are not transparent like window glass. Rather, they are metals with the disordered atomic structure of glass. While common glasses are generally strong, hard, and resistant to permanent deformation, they tend to easily crack or shatter. Metals tend to be tough materials that resist cracking and brittle fracture—but they have limited strength. Metallic glasses, Johnson says, have an exceptional combination of both the strength associated with glass and the toughness of metals.

This video is not supported by your browser at this time.
A piece of metallic glass is heated and smashed in just 10 milliseconds. Credit: Georg Kaltenboeck

To make useful parts from a metallic glass, you need to heat the material until it reaches its glass-transition phase, at about 500 degrees C. The material softens and becomes a thick liquid that can be molded and shaped. In this liquid state, the atoms tend to spontaneously arrange themselves to form crystals. Solid glass is formed when the molten material refreezes into place before its atoms have had enough time to form crystals. By avoiding crystallization, the material keeps its amorphous structure, which is what makes it strong.

A piece of metallic glass being heated and squished in milliseconds, as seen in these infrared snapshots. Credit: Joseph P. Schramm

Common window glass and certain plastics take from minutes to hours—or longer—to crystallize in this molten state, providing ample time for them to be molded, shaped, cooled, and solidified. Metallic glasses, however, crystallize almost immediately once they are heated to the thick-liquid state. Avoiding this rapid crystallization is the main challenge in making metallic-glass parts.

Previously, metallic-glass parts were produced by heating the metal alloy above the melting point of the crystalline phase—typically over 1,000 degrees C. Then, the molten metal is cast into a steel mold, where it cools before crystallizing. But problems arise because the steel molds are usually designed to withstand temperatures of only around 600 degrees C. As a result, the molds have to be frequently replaced, making the process rather expensive. Furthermore, at 1,000 degrees C, the liquid is so fluid that it tends to splash and break up, creating parts with flow defects.

If the solid metallic glass is heated to about 500 degrees C, it reaches the same fluidity that liquid plastic needs to have when it's processed. But it takes time for heat to spread through a metallic glass, and by the time the material reaches the proper temperature throughout, it has already crystallized.

So the researchers tried a new strategy: to heat and process the metallic glass extremely quickly. Johnson's team discovered that, if they were fast enough, they could heat the metallic glass to a liquid state that's fluid enough to be injected into a mold and allowed to freeze—all before it could crystallize.

To heat the material uniformly and rapidly, they used a technique called ohmic heating. The researchers fired a short and intense pulse of electrical current to deliver an energy surpassing 1,000 joules in about 1 millisecond—about one megawatt of power—to heat a small rod of the metallic glass.

The current pulse heats the entire rod—which was 4 millimeters in diameter and 2 centimeters long—at a rate of a million degrees per second. "We uniformly heat the glass at least a thousand times faster than anyone has before," Johnson says. Taking only about half a millisecond to reach the right temperature, the now-softened glass could be injected into a mold and cooled—all in milliseconds. To demonstrate the new method, the researchers heated a metallic-glass rod to about 550 degrees C and then shaped it into a toroid in less than 40 milliseconds. Despite being formed in open air, the molded toroid is free of flow defects and oxidation.

In addition, this process allows researchers to study these materials in their molten states, which was never before possible. For example, by heating the material before it can crystallize, researchers can examine the crystallization process itself on millisecond time scales. The new technique, called rapid discharge forming, has been patented and is being developed for commercialization, Johnson says. In 2010, he and his colleagues started a company, Glassimetal Technology, to commercialize novel alloys using this kind of plastic-forming technology.

Explore further: Physicists advance understanding of transition metal oxides used in electronics

More information: "Beating crystallization in glass-forming metals by millisecond heating and processing," Science, May 13, 2011.

Related Stories

New glass tops steel in strength and toughness

Jan 10, 2011

(PhysOrg.com) -- Glass stronger and tougher than steel? A new type of damage-tolerant metallic glass, demonstrating a strength and toughness beyond that of any known material, has been developed and tested ...

Los Alamos pressure process makes pure zirconium glass

Jul 15, 2004

Zirconium may not be a girl's best friend, but by squeezing the metal with roughly the same pressure needed to make diamonds, scientists at the University of California's Los Alamos National Laboratory made a pure glass that ...

Metallic Glass Yields Secrets Under Pressure

Mar 16, 2010

(PhysOrg.com) -- Metallic glasses are emerging as potentially useful materials at the frontier of materials science research. They combine the advantages and avoid many of the problems of normal metals and ...

Recommended for you

Yellowstone's thermal springs—their colors unveiled

Dec 19, 2014

Researchers at Montana State University and Brandenburg University of Applied Sciences in Germany have created a simple mathematical model based on optical measurements that explains the stunning colors of ...

User comments : 25

Adjust slider to filter visible comments by rank

Display comments: newest first

jscroft
2.4 / 5 (5) May 12, 2011
I have to imagine this stuff would hold an edge pretty well.
TheQuietMan
1 / 5 (5) May 12, 2011
Transparent titanium anyone?
MadLintElf
3 / 5 (1) May 12, 2011
Imagine what you could do with hip replacements, pacemakers, etc.

Oh yea, and imagine a samurai sword, only problem is metallic glass has a bounce factor :-)
that_guy
2.7 / 5 (3) May 12, 2011
I'm not sure it would hold an edge better than contemporary blade making techniques. From my 5 minutes on google, it appears to be possible to make a stronger knife (they already do sell metallic glass knives), but the blade is sharper using regular metal, or ceramic.

For those of you who were confused about how much energy that megawatt is (like I was), 1000 joules is about 1 third of a kilowatt hour - enough energy to keep a 100 watt bulb lit for 3-4 hours, and about 5 cents in electicity.

However, scaled up from a .5cm by 2cm rod, the electricity costs could be enourmous. Consider a .5cm by 10cm by 5cm plate - paying $25 in electricity alone to form a small plate? This won't replace plastic. it will replace metal.

So is the saying going to change from 'they don't make 'em like they used to' to 'they didn't make 'em like they do now' once some applications are changed to liquid metal.

that_guy
not rated yet May 12, 2011
Transparent titanium anyone?


Didn't the article specifically say that this isn't like regular silicon glass in that it's NOT transparent. Did you even read the article??
alec123456789
5 / 5 (1) May 12, 2011
Transparent titanium anyone?

No.
The glassyness is the structure of the atoms. It's kinda like especially thorough work hardening. This stuff still looks like regular metal.
sstritt
3 / 5 (4) May 12, 2011
Transparent titanium anyone?

Apparently no one got your Trek reference to transparent aluminum :(
grgfraiser
not rated yet May 12, 2011
this should be good for the auto industry
that_guy
5 / 5 (1) May 12, 2011
Transparent titanium anyone?

Apparently no one got your Trek reference to transparent aluminum :(

Actually, that's kinda what I thought...but it fell flat when the article explicitly stated the opposite.

Speaking of transparent aluminum - do you prefer glass (Sapphire), crystal (corundum) or ceramic (aluminum oxynitride)?
(and they've been using man made clear sapphire for all sorts of applications for a while now, while the ceramic aluminum is fairly new.)
MIBO
4.5 / 5 (2) May 12, 2011
That Guy,
I think you are still confused.
1 watt is 1 Joule per second, a 100 watt lightbulb consumes 100 Joules in 10 seconds, or 1000 Joules is 1 KilowattSecond, or 1/3600 of a Kilowatt hour.

Actually Ohmic heating is very efficient, probably the most efficient way to heat a metal, so the energy costs of this method are no higher then the normal energy costs of heating metal to it's melting point, in fact probably considerably less.

the maths is not difficult!,
Royale
1 / 5 (1) May 12, 2011
MIBO, What then did they bother using Megawatt in the article for? It's probably just me not making the connection in my head properly, but if you could explain what they mean by '1 megawatt of power' it would be very helpful.
that_guy
5 / 5 (2) May 12, 2011
That Guy,
I think you are still confused.
1 watt is 1 Joule per second, a 100 watt lightbulb consumes 100 Joules in 10 seconds, or 1000 Joules is 1 KilowattSecond, or 1/3600 of a Kilowatt hour.

Actually Ohmic heating is very efficient, probably the most efficient way to heat a metal, so the energy costs of this method are no higher then the normal energy costs of heating metal to it's melting point, in fact probably considerably less.

the maths is not difficult!,

math is not difficult, and my math calculations were not wrong. However, i was calculating based on a wikipedia article on kilowatts that included conversions, and didn't realize that the conversion was listed as a watt hour instead of kilowatt hour...

So you are correct - My results are off by a factor of 1000. Now this technique for molding glass metal seems super awesome.
TheGhostofOtto1923
1 / 5 (2) May 12, 2011
Hey just the stuff for nuclear reactor containment vessels- they just have to raise the melting point-
Vendicar_Decarian
1 / 5 (4) May 12, 2011
Why not try this with George Bush's head and see what happens?

I would personally like to know.
Vendicar_Decarian
1.5 / 5 (4) May 12, 2011
"a 100 watt lightbulb consumes 100 Joules in 10 seconds" - MIBO

FAIL...

100 = 100 * 1 not 100 * 10

Vendicar_Decarian
1 / 5 (4) May 12, 2011
"this should be good for the auto industry." - NoComprehend

Why? The auto industry isn't even interested in making things out of stainless steel. Making things out of glass steels isn't and will never be on their map.

Rust is the automotive Industry's greatest friend.
Vendicar_Decarian
1 / 5 (2) May 12, 2011
"Imagine what you could do with hip replacements, pacemakers, etc." - Whah?

Hip replacements are already strong enough when made from cobalt and chromium steels and are strong enough as is. Although there have been cases of chromium and cobalt poisoning in some patients.
kiwiracer
5 / 5 (1) May 13, 2011
"this should be good for the auto industry." - NoComprehend

Why? The auto industry isn't even interested in making things out of stainless steel. Making things out of glass steels isn't and will never be on their map.

Rust is the automotive Industry's greatest friend.


Actually it will be great for the auto industry. Glass metals will have less issue with warping in extreme heat like steel and less issue with sudden catastrophic failure like ceramic compounds and will therefore be of huge use in brake design, though unlikely to be used in supercars et al since they need a material that gets to working temp quickly (ceramics).
In engine design glass metals will again be in huge demand for weight saving and longevity gains.
Through the rest of the vehicle design carbon composites are still being experimented with and getting cheaper and easier to use all the time. Glass metals are unlikely to be used in chassis or suspension design.
MIBO
5 / 5 (2) May 13, 2011
Vendicar,
well spotted, I missed a zero, maths is not as difficult as typing!.

Royale,
Megawatts ia a unit of POWER, not Energy, they said it takes 1 Milisecond, which is a unit of time, when you multiply Power by Time you get energy.
1MW = 1E+6 J/S
1ms = 1E-3S
1MW * 1ms = 1E+3J
Birger
not rated yet May 13, 2011
Hey! Thermal shields for reusable spacecraft... that do NOT shatter if hit by a piece of foam!!! Maybe something for the russian Kliper, if it ever gets funding.
Royale
1 / 5 (1) May 13, 2011
Got it! Thanks MIBO. I appreciate the response.
Raygunner
not rated yet May 13, 2011
"Hip replacements are already strong enough when made from cobalt and chromium steels and are strong enough as is. Although there have been cases of chromium and cobalt poisoning in some patients.


I have a friend in Atlanta that was subject to a "hip recall" because of the possibility of chromium poisoning. Holy cow - now that's not something you take in and wait in a lobby for! He was in the hospital for 4 days.

Maybe this metallic glass, when coupled with ceramics and other materials (molecular diamond coating?), could be a whole new area in materials science for strength. I doubt it's as lightweight as titanium though.
damnfuct
not rated yet May 13, 2011
First, the usual explanation when talking about glassy metals: glass does not exclusively specify a transparent dielectric material; all it specifies is an amorphous solid. I don't think transparent metal is possible due to definition/properties of a metal (metallic bonds, lack of appreciable band gap).
Second, if I recall correctly, you can cast a glassy metal into a mould and have it come out being sharp. In the case of conventional metals, you'd need an extra step to sharpen the cast materials.
Third, I've heard that the nature of the amorphous metal state provides some protection against rusting (something about grain boundaries).
This sort of news makes me happy. Although glassy metals won't completely replace plastic, there are a few areas that plastics are used only because they are so much easier (cheaper) to produce. Having an easily-cast high-strength metal (cast much like plastic) opens up new and interesting possibilities.
lhc
not rated yet May 14, 2011
Vendicar,
well spotted, I missed a zero, maths is not as difficult as typing!.

Royale,
Megawatts ia a unit of POWER, not Energy, they said it takes 1 Milisecond, which is a unit of time, when you multiply Power by Time you get energy.
1MW = 1E+6 J/S
1ms = 1E-3S
1MW * 1ms = 1E+3J


Or to simplify everyday comparison even more, lets leave joules out.
100W to 1MW is 10000 times weaker so lets multiply time by the same factor:
1MW in 1ms = 100W in 10s
S_Bilderback
not rated yet May 14, 2011
Transparent Aluminum

http://www.physor...273.html

Please sign in to add a comment. Registration is free, and takes less than a minute. Read more

Click here to reset your password.
Sign in to get notified via email when new comments are made.