Tuesday, August 2, 2016

Techy Tuesday - Another 3D Printing Breakthrough

This piece is a couple of weeks old, but interesting and noteworthy.  German precious metals expert Heraeus and Swedish startup Exmet have successfully 3D printed amorphous metals. 

Amorphous metals?  Most people are aware that metals have a crystalline structure.  If you take a piece of wire like a paperclip and bend it back and forth, you increase the size of these crystals through a process called cold-working and it will soon break along these crystal boundaries.  Given a large enough piece being bent until it breaks, these crystals can be easy to see.  Amorphous metals are sometimes called metal glasses because they show the same sort of non-crystalline structures that glass shows.  (amorphous means "without shape")
When I studied Materials Science in college, these were exotic materials.  The only way known to be able to create metal glasses to study was to cool molten metals at extreme rates.  The first experiments chilled the metals at a trillion degrees Kelvin per second.  That's a hard number to grasp, so for example this would be equivalent to cooling molten Nickel at 2913C (3186K) to liquid nitrogen temperatures of 70K in 3.1 nanoseconds.  Only very thin layers (< 50 µm) could possibly transfer heat that fast.  Work since then has focused on coming up with alloys that can be cooled at lower rates, by using metals with extremely different atomic sizes and current numbers for cooling rates are in the range of 1000 K/s and under.  Rates down to 1 K/s have been shown.  This allows much more useful sizes. 

OK, cooling something at a trillion degrees per second is a pretty outrageous achievement, but even 1000 degrees K/sec in a 3D printer sounds extremely difficult.  Why do it?  Heraeus sums it up in this statement:
“Amorphous metals will change our future. They possess a wide variety of previously incompatible characteristics: They are very strong and yet malleable, as well as harder and more corrosion-resistant than conventional metals...”  
Continuing from their press release announcement:
Amorphous metals are suitable for an exceptional number of high-tech applications. They are energy-absorbing and scratch-proof while still having very good spring characteristics – interesting for injection nozzle diaphragms, casing for consumer electronics, or as dome tweeters for speakers. “For fifty years the commercial success of amorphous metals has been held back by inadequate manufacturing methods. Now that changes. Exmet looks forward to cooperating with Heraeus as a competent partner with a worldwide network to help bring this disruptive new technology to the market,” says Mattias Unosson, Exmet co-founder and CEO.
While this was exotic technology when I was in school, today, metal glasses are manufactured in quantity for some specialized applications.  Still, a way to produce them using additive manufacturing appears to be able to open up new applications and new markets for amorphous metals.  It's a shame Heraeus doesn't feel fit to give us even a clue how they do it.  Their release describes what sounds like conventional laser or electron beam sintering of metal powders.  I know it's proprietary, but it would be fun if they mentioned they sintered a spot and then sprayed a drop of liquid nitrogen on it!
They did, however, grace us with this concept artwork.  Exactly what "concept" is being portrayed (other than some random shape coming out of the printer) is a bit beyond me. 


  1. Way back when Scientific American was still a real science magazine, they had an article on amorphous metals, called "Metallic Glasses" back then. I remember a diagram of the machine used to produce them, which was a rapidly spinning copper wheel, cooled by LN2, that the liquid metal was poured (more like squirted under pressure) on to.

    The metal was a special iron/silicon alloy used to make transformer cores with extremely low hysteresis, and thus very little core loss.

    Pretty useful for huge transformers like in power plants, where the core loss can waste significant amounts of energy, and in motors.

  2. Some years ago I read the book, "Venus Equilateral," by science fiction author George O. Smith. In one of the stories he invents the matter duplicator, a device that will break down material and then put it back together to duplicate almost anything.

    3D printing in the the real world continues to break down the wall between science fiction, and science fact.

    Smith also wrote about the far reaching changes to the economy when the widespread use of matter duplicators made working unnecessary.

    1. Essentially the Star Trek replicator - I wonder if that's where they got the idea? They had the advantage of an essentially unlimited energy source with their matter/antimatter reactors and dilithium crystals. Freezing atoms out of energy requires almost unimaginably large amounts of energy.

      In the ST universe, they've mentioned many times that they have no need for money, but never explained how their world worked. I can imagine a time in the future when robots or automation do almost everything, but I can't see how society gets to 5 or 10% employment, with those few people paying for everything.

    2. Well, the answer is essentially this: if everything and anything is free, what do you need money for? The underlying theory is that technology has advanced to the point where anything you want is hanging there for the asking. No-one needs to "work".

      Star Trek explores answers the the question, "What, then, do humans do with their free time?"

      We are slowly getting to that point, and the answer to that question may determine whether humans have a future, or not. It is unfortunate that socialism is becoming intertwined with this, but it need not be.

    3. "...answers TO the question..."
      Dang useless commenting system that won't allow edits! I guess you get what you pay for.

  3. That concept art looks more like something related to the nano tech world. If this involves 3D printing however, the quantities that can be produced are not necessarily small, are they?

    And with those characteristics, I wonder if you can print a sheet and then shape it for other uses? Or does that defeat the purpose of 3D printing?

  4. They likely don't have to rapidly chill. typically amorphous metal products are powder metal to begin with (extemely fine particle size), and are effectively metal-injection-molded / low to medium temperature sintered using HIP or other pressing methods to consolidate without grain growth.
    Since direct metal laser sintering processes when used in thin layers (<50um) have extemely high heating rates (easily 10-100 million-K/s if not higher, given spot sizes comparable to layer thickness, scan speeds of m/s, and delta-T's of 1000+K) and if programmed properly to keep the overall workpiece at a low equillibrium temp, rapid cooling rates (even without active cooling) of at least a few hundred thousand K/s for the initial cooling to get below the grain growth zone.
    So effectively, provided you do it right (and it's likely just picking the right powder size and then programming), AM's should be DMLS-able with simply the right feedstock...which is the naturally produced state of the materials anyway.

    Basically, existing AM stuff is made using lots of pressure but lower temps to consolidate the powders, and this is higher temp, but lower pressure, just very fast.

    I'm actually surprised it has taken this long to marry the two, given that raw AM alloys are powder anyway.

  5. I'm getting a great giggle out of the "degrees Kelvin" here. I know it sounds all sciencey and stuff, but the only difference between kelvin and celsius is where zero degrees is.

    So the rate of cooling is identical in C˚ as it is in K˚...

    1. Yeah, I know. It started out as C/sec, but the references I linked to were all in K/sec, so I went with that for consistency.

    2. A quick note, unlike "degrees C" or "degrees F" the word degree(s) when used with K or Kelvin is unnecessary and technically incorrect. Blah blah K or blah blah Kelvin is the appropriate usage (as our esteemed blog author did correctly, except for once!
      Sorry Angus, sorry SiG, but if we are throwing out technical corrections for a giggle...at least let's be consistent! :)