Podcast Transcript: The Basics of Additively Manufactured (3D Printed) Metals
Release date: September 3, 2020
Nick Birbilis, Deputy Dean of the College of Engineering and Computer Science at the Australian National University in Canberra, recently joined the inaugural episode of the CORROSION journal Interview Series to explain the basics of additively manufactured metals. See below for a complete transcript.
[introductory comments]
Sammy Miles: Nick, thanks for joining.
Nick Birbilis: Thank you so much, Sammy, and thanks to NACE International for having me today. It’s a real pleasure to have the chance to speak to you all.
SM: Thank you again. Before we jump into today’s conversation, do you want to share a few details about your background in corrosion and material science?
NB: I’d love to. One of the things with material science and how I got into it is when I decided to go to college, I had no major. I decided to pick something that was going to be interesting, and of course, the thing with material science is, until you’ve had a little bit of exposure, you don’t realize just how interesting it is. So if you think about some examples of that — the materials on the space shuttle to allow it to reenter and come back to Earth, or the way silicon makes computers work or materials that can harvest electrons — getting excited about material was something that came naturally to me.
I guess the more you dive into materials, the more you realize the engineered products, where humans are responsible for making them and their properties. So then the most exciting part of materials to me is how do you keep them that way? That’s where I got really interested in corrosion in terms of understanding materials and materials preservation, and trying I guess to beat Mother Nature in terms of taking these materials back by corrosion. It’s been a bit of a labor of love, and it continues to be.
SM: Our topic of the day is additive manufacturing, which is also 3D printing in layman’s terms. Generally, what comes to mind is plastics, but there are a variety of other materials that can be used in this process. Could you give a few examples for us?
NB: I certainly can. I’m glad you mentioned that. In layman’s terms it is known as 3D printing, which sounds exciting because it really is exciting. Additive manufacturing captures a whole range of other technologies which don’t immediately strike as a printing technology, but they are additive. In terms of what materials you can use, that’s a really good question. The good news is, nowadays, you can additively manufacturer all the major classes of materials, which is polymers, ceramics, and metals.
Those are the three major classes of materials. There are a few quirks, and I think it’s worth elaborating on what those quirks are. One of the materials that you couldn’t print until very recently, which is a subset of ceramics, was glass. In the last year or so, the technology to 3D print glass has also evolved. There’s a couple of companies that are doing that around the world. One is Micron 3DP. The other one is Maple Glass Printing. I’ll be totally honest. In terms of these classes of materials, you can print the majority of materials within these classes with some exceptions. So there are some trade-offs in terms of additive manufacturing. That relates to limitations from the technology used in additive manufacturing.
An example from metals, which I know is something that corrosion is getting excited about, is that you can 3D print a heck of a lot more materials than you can actually make by conventional methods. But there’s some that you can’t. You can’t 3D print very reactive materials like magnesium or sodium or calcium, which are metals, because the way in which you're doing the additive manufacturing would make them explosive or could be dangerous. Similarly, things like arsenic that create a bit of a vapor. I guess what I’d like to do, if there’s an opportunity to, is elaborate a little bit on how additive manufacturing of metals is done and then understand why some things can’t be additively manufactured but some other things that you couldn’t access before actually can be. That’s where the excitement really lies.
SM: Absolutely. Definitely.
NB: I’ll jump into a little bit of an overview of how the process works with metals. This is going to really test my linguistics because over a podcast, there’s not any visual aid. You’ll have to all use your imagination. So that’s okay. I think you’ve all got really good imaginations, especially if you’re already on this podcast. For 3D printing of metals, there’s two major technologies which make up probably about 90% of the usage at the moment.
The first of those is called selective laser melting, where you have a bed of powder. So basically, you could think of it as a box of metallic powder. The floor of that box can move up and down. What actually happens is a laser hits the powder bed at the interface of the powder and the environment. Where you’ve got the base of that powder bed up really high, you're able to print the layer. Then you drop the powder bed and print another layer and another layer. Actually, 3D printing from selective laser melting is kind of like inkjet printing at home with your printer. It’s a layer-by-layer process where each layer is added to the next one and the melting of that powder locally is done by a very powerful laser. That’s that technology. In that particular case, it’s not a free-form robotic arm that’s moving around and building your part.
Whereas the other technology is called direct energy deposition. That’s the other 3D printing technology that’s very popular, where you basically have a nozzle that has a very large laser beam coming out of that nozzle and a stream of metallic powder is blown into the path of that nozzle. It’s melted in the air or in the inert environment just above the substrate that it’s going to hit. Then you build up by a free-form robotic arm. Those are the two key technologies. Both of them work by having a very high-powered laser. They make up about 90% of the market. But there are other ways of doing additive manufacturing. If you pause about thinking about 3D printing for the moment and just think about how do we have a process that’s additive as opposed to subtractive, where you're taking away material, you can also achieve additive manufacturing by what’s known as wire additive arc melting, which is a little bit like welding. To be quite honest, if you think about welding, if you're welding clumps on top of one another, you can generate a shape with very little material wastage. Wire arc is another process. Another process is electron beam melting, which is much like selective laser melting I spoke about before, but the local heating source is a focused beam of electrons.
The final one is cold spray, so at supersonic speeds firing powder without any laser onto a substrate, means that they collide onto one another and actually bond. The challenge there is that the speed of the powder needs to be blown at a very high speed, so we’re talking supersonic, so faster than the speed of sound. There are commercial systems for all of those. In total, I mentioned six unique processes, but certainly the main two are the laser-based methods. The other ones that have gained in traction, not just worldwide in terms of R&D, but actually the components that are now in a whole range of applications.
Examples include custom titanium implants in humans for hip cups or femurs. Other examples include about 30 different parts for aircraft engines that General Electric is now using. There’s a lot of press all around these. In fact, General Electric bought one of the main companies that produced the selective laser melting machines recently, so they are a producer of not just components but producers of printers to make components as well. Certainly a dynamic time in the additive manufacturing field.
SM: As you mentioned these six different types, what are some of the benefits? Why are we using a variety of techniques? Are some better tailored for certain processes or, for example, component shapes or structure integrity or —?
NB: Yes, that’s a really good question. In a way, the selection of technique is largely known to be informed by what the end use is. That’s a really nice point about additive manufacturing because you can make niche components or bespoke components in small production runs with relative efficiency. What that means is you get to think about the whole of the process. So you get to choose the method that you want to use for your production that matches the end process.
A good example of what might dictate a process selection is something like that final part size. Selective laser melting has one advantage and one disadvantage. The advantage is it has exceptionally good dimensional control, so you can build things with literally micron-sized resolution in terms of contours. You can 3D print directly bolt heads and threads and so forth. But you’re limited to the size of your powder bed. Your powder bed can’t be very large just for practical reasons. Whereas the other process, like direct energy deposition blown powder method, can actually make components that are many feet, many tens of feet in size in each dimension at the loss of extreme dimensional accuracy. There’s some pluses and minuses to each of these.
But in terms of what the benefits are more generally for additive manufacturing, I can deep dive into that a little bit because I think that’s core to what we’re talking about today. Some of the unique benefits, of course, that are really obvious, is the fact that you can produce things in net shape. You can produce them to net final shape. That means no wastage. That’s why the process is additive. Traditionally, what you’d have when you have raw materials or even castings is you have finishing operations that remove material in that the product is not in its final net shape.
By being able to produce things in net shape, what it also means is that you can also build in part complexity. You can build in design freedom that comes for free. And you can also do activities that you couldn’t do in the past. This might sound a little bit sci-fi, but of course, we all probably have thought about if you had a 3D printer on the moon, you could build things on the moon and you wouldn’t need a whole factory there. That reality, toned down a little bit, is you could potentially have a 3D printer on board a ship. You could have a 3D printer in a remote community. You could have one in a hospital. Where if something breaks down or if you need a mission-critical component, you don’t have to wait months and months for it to come from a foundry in Seattle if you’re all the way in Melbourne, Australia. You can produce things locally. So it’s decentralizing manufacturing in a very positive way.
SM: Right. And then all you would need is the metal powders or whatever components you have to go into it to make it right? So you would have that on hand versus needed to order.
NB: That’s a really good point. You’d have that in stock just like you’d have it in your pantry ready to go. That is a good point.
SM: On most of the components, are you just using a single metal? Are you blending together different powders to make that end product that you want?
NB: Good question, Sammy. Generally, you can do both, and it depends what it is that you want to make. So this is something where I can babble on for ages when it comes to the alloy design and the alloy considerations. The key thing is the different technologies give you different flexibility in being able to alloy on the fly. Powder bed method like selective laser melting means that the composition of your powder bed is usually fixed before you start printing, but you can blend powders in there and generate unique alloys. That’s in fact what a lot of researchers are doing at the moment. So creating using unique alloys or mixing powder that have not been mixed before.
Whereas direct energy deposition, because you’re blowing powder into a stream, you can be blowing in from multiple hoppers, and you get the ability to produce composite materials or hybrid materials or mixing metals and non-metals at the same time. So having ceramic-metal composites, for example, using waste material in your feedstock. It is a good question. You do have that metallurgical freedom.
But your question reminded me of the one important thing that I wanted to mention here. When I mentioned earlier you have some limitations with laser-based methods, like you can’t 3D print explosive materials because the interaction with the laser gets to a very high temperature and things will go POW. What you can do, however, is for materials that are not so reactive, which includes the transition metals, which includes the huge chunk of materials in the middle of the periodic table — so we’re talking about dozens of metallic elements in the middle of the periodic table — can now be alloyed with one another. That’s part of the really big excitement about additive manufacturing of metals, where in a way, those in the industry are now calling it a bit of a renaissance in metallurgy because we have great opportunities to create a vast number of alloys that we couldn’t access before.
For those that don’t know much about alloy production and how alloys are made, whether it’s stainless steel or aluminum alloys, normally there’s a process where you go from liquid and you cast into some sort of ingot at some stage, where you go from a liquid to a solid. That’s usually governed by mutual solubility of one element in the other. It’s kind of a bit like the oil in water issue. You can’t make alloys where you get phase separation, where you have like a separation of oil and water. You need to work within mutual solubility of materials. In the case of mixing materials under a very high-powered laser, you have such an intense power going into that local region that everything is annihilated quickly, and then it’s rapidly solidified when the laser moves position to go and print another bit of your component. You get rapid heating and rapid cooling that you're basically cheating nature, in a way, and able to create alloys that have never been seen before.
This is really exciting. I can’t overstate that. With that level of excitement, of course, comes some challenges. One of them is that, what do you make now that you can create billions and billions of different compositions? People will be working in this area for a very long time but for very good reason, I think.
SM: That is exciting to be able to create new materials, blend new items together. What type of impact does that have on the actual properties, though? If you can blend new alloys and let’s say have this whole new renaissance going, are you expecting — can you all predict what the structure’s going to be, how it will hold up, how it will last long-term, or the corrosion properties, anything like that?
NB: These are all really good questions.
SM: Is it a little too early to know?
NB: In a way it is. But these are the important questions to be asking right now. And these are the important questions to be putting forth to the corrosion community because we have the opportunity to learn lessons from the past, where a normal alloy development took upwards of decades, many decades put together if working in isolation. Now with better communication, with better characterization, we have the ability as a community to move the dial quicker.
When you mentioned — I’ll try to keep it high level — but when you mentioned what type of impact can the process have on the properties, inherent to 3D printing — and it’s regardless of all the methods that I mentioned earlier — to some extend you get rapid solidification and you get that to a large extent with laser-based processes. So what it means is you're developing structures that are basically frozen in time in what may be a so-called metastable state. So how their long-term stability plays out is something that’s yet to be seen. I mean, in cases for the very long term. I don’t mean days or weeks, I mean in the very, very long term. But it also means that we can start to ask fundamental questions.
Like I alluded to earlier, metals, which is largely what we make in our structural materials and engineered components with, in their own right, they’re already a human-made endeavor. You don’t just walk around the forest and find big chunks of metal. Usually they’re refined from ores and minerals and have a lot of energy put into them. Metals themselves are a human-made construct that are largely metastable regardless. It gives us a philosophical way of, a new philosophy in the way we design materials.
The good news, though, is the properties so far have been considered to be quite excellent for 3D printed components. I’ll say that with a little star next to it. What I mean by excellent, I mean the rapid solidification usually means that you can get fine-grained structures and that you get good chemical homogeneity through components. So there are some real benefits, and that translates to higher strength. But if process control starts to become an issue, then you may have defects in there, which deteriorate properties as well.
I might just elaborate a little bit further on some of the challenges when studying additive manufacturing, additively manufactured materials, because processes will be part of that. If it’s okay, Sammy, I’ll elaborate a little bit on some of the challenges.
SM: Absolutely.
NB: One of the things when you're 3D printing metals — and if you google this and look at some of the research papers — is there’s a few key challenges that are evident. One is the reproducibility from different instruments, in different countries or different states or different continents. So the reproducibility of properties from instrument to instrument.
What that means is a researcher or factory in one country may use a certain laser power, and in another factory, they may use a different laser power. Both of them make a solid material that looks the same, feels the same, and has the same application but has different properties. Fortunately, in the last couple of years, the not-for-profit entity in the U.S. called ASTM has looked at implementing standardization processes across additive manufacturing, which is of course really useful. Reproducibility remains a challenge.
But of course, the other challenge — and this is where, finally, we get to talk a little bit about corrosion — is defects and defect control. What would be considered a satisfactory build of an additively manufactured component is if you can build it with the minimum amount of defects towards a negligible amount of defects. I’m talking about inclusions, oxides that might come from interactions with gas or poor environment control, or even porosity, and there’s various types of porosity. Minimizing those is important for defect control. If you can’t minimize those, one of the challenges out there is what effect is that having on corrosion and do we understand that effect? I think it’s probably fair to fall on our sword as corrosion engineers and say we don’t understand completely at the moment the effects on corrosion of defects that can occur in additively manufactured materials.
To pick one example, if you have a material that’s printed and not 100% dense — it would be 99.5% dense and have some porosity throughout the material — what will that do to its long-term corrosion behavior? Research to date has indicated that it actually has perhaps minimal — fortunately, minimal to negligible — effects on corrosion initiation. But what research is also showing is once corrosion has initiated, it’s much more difficult to repassivate such materials.
Getting to that level of understanding for our field would be important before 3D printed components ultimately go for vast mass-market type applications, in the as-printed condition. I should say that there’s a lot of clever people out there that have thought about this, of course, and there are ways of avoiding — I wouldn’t say avoiding — correcting defects in 3D printed components after you have produced them. So there are various heat treatments or, of course, isostatic pressing, which is what can be used to provide heat treatment and pressure into a component after you’ve built it such that you minimize or eliminate the porosity. That’s typically what’s done to components that are going to end up flying or ultra-critical components that may be subject to repetitive fatigue loading and so forth.
But the holy grail would be able to have a printer somewhere, tell it what you need, it makes it, and then you take it out, and then you use it. We’re not there just yet for a lot of engineered components, but we are for a good number of them.
SM: That’s promising, and it sounds like there’s a lot of areas for future research to continue in the field, both in terms of building the alloys, in terms of how they last, corrosion properties, things like that. We’re getting near to the end of our time. Do you have any additional things you’d like to say before we wrap up?
NB: Yes, Sammy, I think this is a really hot topic, and I’m delighted to have been the one to be able to talk about it. Clearly there’s a lot of great researchers, both corporate and academic, that are working on this at the moment. I’d encourage the field to communicate and talk to one another and really share experiences, share knowledge, because I don’t want to overstate it, but materials really are getting critically important in the next 10, 20, and 50 years.
So if we think about some of the grand challenges that we have facing our societies — things like clean energy, clean water, things like cheap air travel — all of these aspects — or personalized medicine — all of these aspects can have a step change in their deployment through additively manufactured metals. It really is that serious. So I encourage people to keep an open dialogue with one another and engage with one another and certainly engage with those involved in NACE. I’d be delighted to connect with as many listeners as possible on this topic.
SM: Thanks, Nick. On that note, what’s the best place for them to go if they want to learn more about the research you're doing or if they want to connect with you?
NB: I’m very responsive to email, so it’s nickbirbilis@anu.edu.au. You can find me pretty easily because I think accompanying this podcast it will have my name, so if you pop that in a web browser, you’ll be able to find me through your favorite social media platform. Of course, I think it would be only fitting to continue the dialogue through NACE as well, and their various platforms, because this is really a hot topic and exciting. I can’t wait to be able to collect and share more response on this topic, as I know other researchers in this field are also itching to do.
SM: Thanks again, Nick. For listeners who want to learn more about corrosion of additively manufactured metals and alloys, Nick was actually one of the co-authors of a review article we published in Corrosion Journal in December 2018. You can find that on our website at www.corrosionjournal.org.
[closing statements]