Printing the metals of the future

Jul 29, 2014 by Elizabeth Landau
Scientists make a rocket nozzle using a new 3-D printing technique that allows for multiple metallic properties in the same object. Credit: NASA-JPL/Caltech

3-D printers can create all kinds of things, from eyeglasses to implantable medical devices, straight from a computer model and without the need for molds. But for making spacecraft, engineers sometimes need custom parts that traditional manufacturing techniques and standard 3-D printers can't create, because they need to have the properties of multiple metals. Now, researchers at NASA's Jet Propulsion Laboratory in Pasadena, California, are implementing a printing process that transitions from one metal or alloy to another in a single object.

"You can have a continuous transition from alloy to alloy to alloy, and you can study a wide range of potential alloys," said R. Peter Dillon, a technologist at JPL. "We think it's going to change materials research in the future."

Although gradient alloys have been created in the past in research and development settings, this is the first time these have been used in making objects, such as a mount for a mirror, said John Paul Borgonia, a JPL mechanical engineer.

Why would you need to make a machine part like this? Say you want a object where you would like the ends to have different properties. One side could have a high melting temperature and the other a low density, or one side could be magnetic and the other not. Of course, you could separately make both halves of the object from their respective metals and then weld them together. But the weld itself may be brittle, so that your new object might fall apart under stress. That's not a good idea if you are constructing an interplanetary spacecraft, for example, which cannot be fixed once it is deployed.

This is a prototype of a mirror mount that scientists made using a new 3-D printing technique. Credit: NASA-JPL/Caltech

JPL scientists have been developing a technique to address this problem since 2010. An effort to improve the methods of combining parts made of different materials in NASA's Mars Science Laboratory mission, which safely landed the Curiosity rover on the Red Planet in 2012, inspired a project to 3-D print components with multiple alloy compositions.

Researchers from JPL, the California Institute of Technology, Pasadena, and Pennsylvania State University, University Park, joined forces to tackle the issue. The result has implications for space travel and machinery on our own planet.

"We're taking a standard 3-D printing process and combining the ability to change the that the part is being built with on the fly," said Douglas Hofmann, a researcher in material science and metallurgy at JPL, and visiting associate at Caltech. "You can constantly be changing the composition of the material."

In their new technique, Hofmann and his colleagues deposit layers of metal on a rotating rod, thus transitioning metals from the inside out, rather than adding layers from bottom to top, as in the more traditional 3-D printing technique. A laser melts metal powder to create the layers.

Future space missions may incorporate parts made with this technique. The auto industry and the commercial aerospace industry may also find it useful, Hofmann said.

A report on this work was published in Scientific Reports on June 19.

Explore further: Soft metal weld makes for stronger infrastructure

add to favorites email to friend print save as pdf

Related Stories

Probing metal solidification nondestructively

Apr 14, 2014

(Phys.org) —Los Alamos researchers and collaborators have used nondestructive imaging techniques to study the solidification of metal alloy samples. The team used complementary methods of proton radiography ...

Japanese material scientists develop new superelastic alloy

Jul 01, 2011

(PhysOrg.com) -- Working out of Tokyo University, scientists in the Department of Materials Science, have developed a new metal alloy that unlike other “superelastic” alloys can resume its original shape in temperatures ...

Glasses strong as steel: A fast way to find the best

Apr 13, 2014

Scientists at Yale University have devised a dramatically faster way of identifying and characterizing complex alloys known as bulk metallic glasses (BMGs), a versatile type of pliable glass that's stronger than steel.

Recommended for you

Using sound to picture the world in a new way

10 hours ago

Have you ever thought about using acoustics to collect data? The EAR-IT project has explored this possibility with various pioneering applications that impact on our daily lives. Monitoring traffic density ...

Sweeping air devices for greener planes

Oct 21, 2014

The large amount of jet fuel required to fly an airplane from point A to point B can have negative impacts on the environment and—as higher fuel costs contribute to rising ticket prices—a traveler's wallet. ...

User comments : 21

Adjust slider to filter visible comments by rank

Display comments: newest first

antialias_physorg
5 / 5 (8) Jul 29, 2014
A strange result of technologoies like this might be to change the way we think. We, as a technology/tool using species are currently used to thinking in terms of assemblies: Solving problems (be they engineering or cognitive) by taking them apart and thinking about each part separately...then assembling the whole into a solution.

When technology like the above matures we may start to think holistcally: interms of transitions from one proble/property space into another.

Or we might not. In any case: It looks like future products won't have welds and screws and nuts and bolts anymore. Just one continuous object with the desired properties it needs to have at each place.
Arties
5 / 5 (4) Jul 29, 2014
rockwolf1000
3 / 5 (2) Jul 30, 2014
A strange result of technologoies like this might be to change the way we think. We, as a technology/tool using species are currently used to thinking in terms of assemblies: Solving problems (be they engineering or cognitive) by taking them apart and thinking about each part separately...then assembling the whole into a solution.

When technology like the above matures we may start to think holistcally: interms of transitions from one proble/property space into another.

Or we might not. In any case: It looks like future products won't have welds and screws and nuts and bolts anymore. Just one continuous object with the desired properties it needs to have at each place.


That would be great for corporations. Cars and trucks with no nuts and bolts. If a part breaks, just throw the whole thing in the trash because it can't be repaired.

Our throw away society takes another leap forward.
Dr_toad
Jul 30, 2014
This comment has been removed by a moderator.
antialias_physorg
5 / 5 (4) Jul 30, 2014
just throw the whole thing in the trash because it can't be repaired.

Why? Just cut the affected part out and put the entire thing back in the printer and fuse a new section back in using the same process that you used to build it.

There isn't a limit to *when* you can do additive metal printing to an object, is there?

Certainly would be great for such 'repair' shops in other ways: they don't need to have warehouses of spare parts lying around (which they may never sell and therfore have to add to the price of other services/goods in order to make sure they recoup the cost).
They just need a bunch of containers with the raw materials, a big printer, and the file of the object to be repaired.

And taking it a step further: Think about the reverse process. UN-printing such an object: melt it apart piece by piece and reuse the material. No waste ever again (and no need to have more than one of anything - which'd significantly reduce clutter in most homes)

Dr_toad
Jul 30, 2014
This comment has been removed by a moderator.
rockwolf1000
5 / 5 (1) Jul 30, 2014
Why? Just cut the affected part out and put the entire thing back in the printer and fuse a new section back in using the same process that you used to build it.


Just a little problem called logistics. Ever worked on a car? What tool would you use to cut off a cracked exhaust manifold with umpteen parts in the way and near impossible visibility let alone access. Then ship the car to where to get the re-printing done? Far easier to unbolt the old and re-bolt the new.

There isn't a limit to *when* you can do additive metal printing to an object, is there?


Perhaps when you need a repair in location "A" and the printer is in location "B" 7500 km's away?

Certainly would be great for such 'repair' shops in other ways: they don't need to have warehouses of spare parts lying around (which they may never sell and therfore have to add to the price of other services/goods in order to make sure they recoup the cost).


Again accessibility is the issue.
rockwolf1000
5 / 5 (1) Jul 30, 2014
@A_P
Don't get me wrong. I love 3D printers. I used to have one in my office to make plastic prototypes. They will allow the manufacture of things not possible or cost prohibitive with conventional manufacturing technology.

However I'm not convinced that making large structures as a single continuous part is a prudent strategy for the reasons mentions above.

It's always easy to just say cut the damaged part away but much more difficult in practice.

Mechanical cutting (saw, water jet, etc.) may not be possible for various reasons.

EDM wire cutting not always practical or possible.

Thermal cutting (plasma, torch, laser) may not be possible or desirable or safe.

How do you propose to cut away a part if none of the above possibilities are suitable?

The screw thread is probably one of the most important inventions of all time and I doubt it will disappear anytime soon.

Dr_toad
Jul 30, 2014
This comment has been removed by a moderator.
antialias_physorg
5 / 5 (1) Jul 30, 2014
What tool would you use to cut off a cracked exhaust manifold with umpteen parts in the way and near impossible visibility

Since you're going to be printing you need not cut a small part. The simples possible way would be to take the object, mount it overhead, and then melt your way in in a semicircular way until you reach the broken part (who cares if you melt away working parts in the process?). Turn around and print it back up. Full access. No expertise whatsoever required (you could even fully automate the process with minimal hardware/software requirements)

If you want to minimize work then an endoscopic approach would be the way I'd go. After all: printing/un-printing is a lot like working on biologicals. Why not take a pointer from surgery methods?

is in location "B" 7500 km's away?

Since such a printer is WAY cheaper than the collection of spare parts and tools found at any garage: why would you think they'd be spaced so far apart?

antialias_physorg
5 / 5 (1) Jul 30, 2014
double post
antialias_physorg
5 / 5 (1) Jul 30, 2014
Don't get me wrong. I love 3D printers.

I know. Such exchanges are important. You only ever know if something is good if you throw every criticism possible at it and see if it could be handled.

I agree that screws and nails will still be with us for some time, yet. But especially when we think off-world then printing unified structures makes a hell of a lot of sense. You just can't build factories for every small replacement part that might be needed on the Moon or Mars (much less in orbit).

And I don't see it making less sense down here. It's also a liberation from proprietary designs, where companies (and entire industries in collusion) charge you more because they are the only ones who have the factories - not because their product is actually worth the price.
The open source community for 3D printing is already producing some amazing designs.
rockwolf1000
5 / 5 (1) Jul 30, 2014
The simples possible way would be to take the object, mount it overhead, and then melt your way in in a semicircular way until you reach the broken part


I'm not sure I'm following you here.

(who cares if you melt away working parts in the process?).


Umm, I would probably care if they cut my fuel and brake lines in order to fix my manifold.

Just go look under any car's hood. Try to find the exhaust manifold. Now imagine the engine block and the manifold are one continuous part. How would you remove and re-print the manifold with all the wires, hoses, belts, sheet metal and other parts in the way?

If heat was used, how would you protect the other parts from damage? or prevent the adulteration of the metal via gases? or warping or shrinkage?
(Some metals shrink when heated to high temperatures)

Remember, it should only take a few hours to re&re the manifold the old way. How long would your process take. At what cost? I only need a socket set to do it. $20
rockwolf1000
5 / 5 (1) Jul 30, 2014
It's also a liberation from proprietary designs, where companies (and entire industries in collusion) charge you more because they are the only ones who have the factories - not because their product is actually worth the price.


Sorry but processes are proprietary these days too.

For instance the oil and gas industry uses special threads on the drilling rods and equipment. Some situations call for a special thread which could be made practically anywhere but the thread geometry and the thread cutting process are licensed to certain machine shops only. Yes they are expensive.

As a machinist and a CNC programmer I could easily make this thread but it would be illegal and the customer won't accept it without the "certificate". Meanwhile we can make dozens of other thread types to high tolerances with no legal issues.

I expect this situation to become more common.
antialias_physorg
5 / 5 (1) Jul 31, 2014
Just go look under any car's hood. Try to find the exhaust manifold. Now imagine the engine block and the manifold are one continuous part.

I own a compact car (smart roadster). The official procedure for changing the bottom 3 sparkplugs is to take out the entire motor.

How would you remove and re-print the manifold with all the wires, hoses, belts, sheet metal and other parts in the way?

I'd come in from the other side.
The cool thing about hoses and the like is: you don't need them anymore. You can print appropriate hollow spaces in the material (That's why I'm excited about printing houses. You could print all the duct works as holow spaces in the same go and the only case you would ever have to worry about leaking pipes is when someone explosively dismantles your house).

Printing wires with metal printers is trivial (printing insulation at the same time is not).
antialias_physorg
5 / 5 (1) Jul 31, 2014
If heat was used, how would you protect the other parts from damage?

Same as when you print: you do it with pinpoint accuracy and one tiny bit at at time (not ethat during printing parts that are printed in layers below the one you are working on don't melt again, either). Since your object will be (almost) upside down for this the molten metal will not run and melt other parts.

Umm, I would probably care if they cut my fuel and brake lines in order to fix my manifold

EVs? No fuel line (and no brake line, either if you don't want to.) I don't see a big future for gasoline powered cars. At least not in the timeframe it would take to switch over to integrated print-manufacturing on a large scale.

Remember, it should only take a few hours to re&re the manifold the old way. How long would your process take. At what cost?

Depends on the extent of damage. Cost would be low (no spare parts - only raw materials, no labor cost - only energy costs)
antialias_physorg
5 / 5 (1) Jul 31, 2014
Some situations call for a special thread which could be made practically anywhere but the thread geometry and the thread cutting process are licensed to certain machine shops only.

Yes. The amount of different screws used in the medical sector alone is amazing. (I have here a 500 page catalog...almost exclusively different types of screws. Prices per item like you wouldn't believe.).
In medical applications the reason why stuff is expensive is the same: The certificate/testing

But integrated/free flow design is hard to patent. You can't just patent "shoe" and expect thereby to foil anyone from printing their shoe design (and patents eventually do run out). I've already seen printed casts (which look really cool)
And most applications for everyday use don't require stuff that is *that* specialized.
lankester
not rated yet Jul 31, 2014
I wonder if similar principles could be used to de-construct 3D-printed items. Disassembling them bit by bit, layer by layer, thus recovering original material to be re-used later. If, for instance, laser melting is used in printing to fuse particles together to form an object, could it also be used on an existing object to re-heat small bits and remove them, e.g. by some sort of a precise vacuum cleaner and then store them as raw material for printing something else later?

That would be a nice way to recycle objects. You could easily and cheaply upgrade whenever an updated model is available, by re-using the original material.
antialias_physorg
5 / 5 (1) Jul 31, 2014
I wonder if similar principles could be used to de-construct 3D-printed items. Disassembling them bit by bit, layer by layer, thus recovering original material to be re-used later.

That would work if the material is pure or at the very least of a homogeneous alloy. If it's heterogeneous (as described in the article) you will have to just melt it down and separate the materials out.
While that is an energy intensive process it's very well doable (and being done on industrial scales all over the world).

Household separation of materials is not yet available. But I wouldn't bet on this state of affairs staying that way for long.
lankester
not rated yet Jul 31, 2014
I wonder if similar principles could be used to de-construct 3D-printed items. Disassembling them bit by bit, layer by layer, thus recovering original material to be re-used later.

That would work if the material is pure or at the very least of a homogeneous alloy. If it's heterogeneous (as described in the article) you will have to just melt it down and separate the materials out.


It should be doable in theory - if the printable file contained some code (QR code perhaps) with information where to find detailed specifications of the printed object and then imprinting it during the process, all that would be necessary is for the disassembler to read the code, download the specs and it would know precisely which location contains which material. Then the separation could be done automatically while disassembling the item.

I'm just thinking out loud here, wondering if such a thing would actually work :-)
antialias_physorg
5 / 5 (1) Jul 31, 2014
download the specs and it would know precisely which location contains which material.

As noted: for homogeneous material: yes.

But picture something like described in the article: An alloy to alloy transition that has 10% X and 90% Y composition at one end and 90% X and 10% Y composition at the other end with a gradual change in between.
Each point you melt will have a different composition. You could catch all of that and get the intermediate 50/50 alloy. But that won't help you if you want to print the same part again. For that you'll have to separate the materials out the traditional way and then use those to reprint.

if the printable file contained some code (QR code perhaps)

STL files are tesselations of your object geometry. Some extensions can already be tagged with additional properties. (STL stands for 'stereolithography'...which basically was the first 3D printing process used on any scale)
rockwolf1000
not rated yet Jul 31, 2014
@A_P

I do see your point, however I do think that keeping things modular is the best route. It simply gives you more options for repairs or upgrades. Options that simply aren't available with large continuous structures.

What I'm confused about is how they get an alloy to have the correct crystal structure after printing. Tool steels for instance arrive at a machine shop in an annealed state so it can be worked. One the shape is finalized, the part has to go for hardening and tempering which is a fairly demanding process in that the material needs to brought to a certain temperature in a vacuum or molten salt bath then cooled at a known rate the reheated to a lower temperature then re-cooled. With some parts we used to request a double draw to ensure the correct temper. I can not see how they can control the rate of cooling with this process so it would seem a secondary heat treating process would be required. This almost always adds some degree of distortion or warping to the part.
rockwolf1000
not rated yet Jul 31, 2014
cont'd

If the material does require some type of heat treatment process. This would probably necessitate the normalization of the whole structure possibly before or after the repair, before the appropriate heat treating process could be applied.

Remember, these heat treatment processes usually must occur in an inert or vacuous atmosphere to prevent oxidation etc. (sometimes a stainless steel foil wrap will work)

Heat costs money and has other environmental costs so avoiding having to normalize and re-heat treat a large structure would be prudent.
antialias_physorg
5 / 5 (1) Jul 31, 2014
I do think that keeping things modular is the best route.

To a point. Anything where you have wear and tear (e.g. tires on a car...though that doesn't mean all moving parts. I could well see axels or electric motors or batteries) being printed as integral parts. Because the chance of failure over the lifetime of a vehicle is low.

Anything static or non-serviceable (like houses, bodywork for cars/planes/ships ..and the entirety of a spacecraft) doesn't need to be modular, and probably would only suffer from being modular over being an integrated whole.

I can not see how they can control the rate of cooling with this process

Just guessing: but by adjusting printing speed/and printing energy it would seem plausible to me that you can locally manufacture the desired hardening/annealing effects without having to do this 'in bulk'. Which certainly would cut down or totally eliminate bulk effects (warping, shrinking, etc. )