PAEK: The Superhero of Ultra-High-Performance Materials

Carl Douglass (00:00):

If PEEK is a superhero, PAEK is like the true superhero because not only does it serve these critical applications, but it can actually get there.

Robert McKay (00:10):

There’s over 20,000 aircraft today that fly with PEEK components in them, something like 13 million implants on the medical side that have been implanted, that are based on our materials.

Silvia Berretta (00:21):

We change the crystallization speed, so it crystallizes much more slowly than PEEK. Widen your processing windows.

Brian Douglass (00:29):

The geometries can be produced relatively easy with LMPAEK and this is really exciting for us.

Carl Douglass (00:36):

Solutionology is about being unyielding with perseverance to get to the solution.

Speaker 5 (00:42):

To not give up and to constantly drive for better. So even when we deliver 100% percent, I want to deliver 110 next time.

Brian Douglass (00:49):

And for me, the constraints of that project are the most important because that’s what drives us to a solution. It’s all about painting a picture and getting all the details in.

Carl Douglass (00:56):

How do we develop a tool that helps share our journey, educate others, and bring more light to the realities of additive manufacturing? That’s Solutionology podcast.

(01:11):

Welcome to the Solutionology podcast, and today we’ve got a couple of really great guests. We’re going to be talking about ultra-high performance materials, and it’s a topic that is really important to us because so much of our work is centered around solving problems with demanding applications requiring ultra-high performance materials, and it’s an area that we spend quite a bit of time, obviously, in fused filament fabrication printing. So that’s what we’re going to talk about today. This episode is really for a very specific audience. We’re going to start talking about PEEK, which is polyetheretherketone. I always have to remember that. Polyetheretherketone. And it’s a material that is printed with FDM and it enables really high temperature, high performance applications to be successful. And it’s a material that’s also very difficult to print with.

(02:16):

So if you’re familiar with the term PEEK, then this conversation is going to resonate with you. And if you’re not familiar with it, you might still want to stay tuned in because it’s likely that especially if you’re in an industrial application, there are aspects of your business or applications that could require PEEK and you just don’t know about it. It is not talked about enough because it is so challenging to work with. However, it is used in machining and molding applications heavily in aerospace and healthcare industries. The way that we look at PEEK is it’s a superhero. It’s able to withstand these high temperatures. It has excellent mechanical properties and it performs really well in harsh chemical environments. And we’re using it in healthcare, aerospace, defense applications, electronics, and heavy industrial. But the challenge for us, and Brian, you can speak to this better than I can, is that it’s really tough to process.

Brian Douglass (03:25):

PEEK has a very tight window for getting strong parts, parts that don’t deaminate, and there’s certain geometries that work well with PEEK, but they’re very specialized. Setting up a process for PEEK requires a tremendous amount of time to make sure that part is going to come out. From a volume standpoint, each part is going to be the same and it’s very repeatable. So PEEK is really challenging to work with fuse filament side for successful parts.

Carl Douglass (03:57):

So most of our PEEK production is truly production where we’re establishing a baseline profile and then producing those same parts time and time again, right?

Brian Douglass (04:09):

Yeah. We’re not producing prototypes unless there’s a very high value prototype that has to be done because of that setup and preparation time on the process controls, it’s so unique to each geometry that it’s not a plug-and-play solution.

Carl Douglass (04:28):

So it’s not that it can’t be done, it’s that the economics oftentimes don’t work out for those geometries.

Brian Douglass (04:35):

Yeah, the demand for the material properties, it really has to be there to justify PEEK for an application. And in most cases, we can find other materials that solve that problem and meet the requirements, it’s just working through that process of education and ensuring that we are hitting the requirements for that project.

Carl Douglass (05:01):

I know that every time I’m talking to a client or working on a project and I go and identify an opportunity for PEEK and I get an opportunity to talk to you or talk to the production team, I’m finding faces that are like… In fact, I don’t even bring them up anymore. Sometimes I’m afraid I’m going to get taken out back and… Taken to the woodshed. But the problem is is that those applications still exist. And that’s why this conversation today is so critical and why I am so excited about it because we’re here with Robert McKay and Silvia Beretta, both with Victrex, which is a chemical company, and interestingly enough, Victrex invented PEEK back in the 1980s.

(05:55):

So there’s a lot of history that we’ve got with the Robert and Silvia here today. And taking that one step further, this is where it gets really exciting, is that we’re going to be talking about PAEK, which is polyaryletherketone. I probably pronounced that wrong and that’s okay. But we’re going to be talking about PAEK. If PEEK is a superhero, PAEK is like the true superhero because not only does it serve these critical applications, but it can actually get there.

(06:28):

So Robert McKay is head of new business development and for additive manufacturing, and you’ve been in this role for over eight years with Victrex, but you’ve got a really storied career history in additive manufacturing, and more importantly, high, ultra-high performance polymers. So there’s a lot of experience that you’ve amassed over the years. Robert, we’re excited to hear your feedback and perspective. And Silvia, Dr. Silvia Berretta, PhD, you are focused on AM technology strategy or as your title has it, AM Strategic technology manager. And you’ve been in this role for over five years with Victrex, and you also have a very strong background in academic and applied sciences for polymer science and engineering. So this is a pretty special conversation because we’re leveling up the discussion beyond what’s the processing window of these polymers and talking about some pretty special things.

(07:37):

Thanks so much for joining us, Robert and Silvia. If you wouldn’t mind just sharing a little bit about your background in Victrex, that would be really great.

Silvia Berretta (07:47):

I can go first. Yeah, my background is in… I started as a researcher in [inaudible 00:07:56] of PEEK and then worked with other high temperature polymers. And then when I joined Victrex, I worked with PEEK with the traditional manufacturing processes across a variety of applications ranging from really consumer goods, manufacturing, up to implants, medical implants. And then in the last two years, now I’m back on focusing on AM in Victrex. Yeah. Rob?

Robert McKay (08:23):

Yeah, and thanks for the introduction, Carl. It’s not many times that we get associated with a superhero material and a storied career, so thank you for that introduction.

Silvia Berretta (08:33):

It’s true.

Robert McKay (08:33):

Yeah, so I’ve been in the industry in high performance polymers and engineering polymers for almost 25 years, I think. It feels like now 22, somewhere around there. I’m losing track. I started with SABIC and was in the polyetherimide, their trade name Ultem business, so related to the high performance polymers there. And then as you said, around eight years in Victrex [inaudible 00:08:55] to manufacturing. I’m a bit of a maker myself and love… As we were talking about before, we started the ME background, mechanical engineering background, we love to tinker and I’ve got an Ender-3 in the basement all hacked out and do a little PLA and ABS printing myself. So it gives a little bit of different color to the job in terms of trying to add it to manufacturer with something like PAEK and having some empathy for what it takes to make it happen, having to play around on my own projects.

(09:25):

So that’s a little bit about me. If it’s all right, Silvia, I’ll start with a little background on Victrex to build on what Carl’s already mentioned. As you mentioned, Carl, 40 years in PEEK, you can blame us for all those spaces you get for the fact that the material exists and that it is specified and it works in so many applications. The original PEEK, so polyetheretherketone was invented for injection molding. And injection molding is a very fast process. You want to make thousands of parts, cycle times in seconds. It’s a very different process than additive manufacturing and printing. So it’s not surprising that trying to take PEEK and put it into your machines that it’s really hard to get to work because it wasn’t designed to work in an additive manufacturing process.

(10:23):

You mentioned some of the industries, I won’t go over them again. You hit them all for the most part that we focus on. Just some evidence of PEEK’s use in the worlds. There’s over 20,000 aircraft today that fly with PEEK components in them. There’s something like 13 million implants on the medical side that have been implanted that are based on our materials. So it’s got a long track record and a lot of trust that’s built up around it. And that’s one of the reasons that when customers come to you and say, “I want additive manufacturer at PEEK,” and then you say, “Let’s use something else instead,” they get a little nervous because they really trust it. And that’s the challenge that we have with how can we make additive manufacturing easier while still giving our mutual customers the products and the performance that they trust. And I think we’ll talk a lot more about that as we go through this discussion.

(11:24):

One last thing I’ll mention before I turn it back over to Silvia on some of our additive stuff. A little known about Victrex is that we don’t just make PEEK polymers, we also use them to make things. So about three hours east of you guys in Grantsburg, Wisconsin, we have a gear manufacturing facility where we make high performance gears. And then in Rhode Island here in the US we have a composites aerospace component manufacturing company where we make aerospace brackets and components with these materials that then fly. And so we manufacture things too and we want to use that to manufacturing ourselves. So we’re on both ends of this supply chain and we have to live the pain of trying to add it to manufacture parts and components with our own materials, and that gives us a little bit more empathy as well into the challenges that you and your customers face.

Silvia Berretta (12:20):

Yeah, and maybe just to elaborate a bit more how the AM world is in Victrex. So we have what we call an ecosystem. So we have different collaborators, partner we work with. We have a joint additive manufacturing center at the University of Exeter in England, southwest of England, if you know where London is. It’s basically three hours drive west from London, which for American standards, it’s very, very short drive, not so much for European standards. And in this sense, for example, are located many filament fusion machines for high temperature polymers, laser sintering machines for high temperature polymers. And then we work with other partners and resellers and other research institute to work on different parts, areas of the AM journey really, looking at different angles modeling or post-processing or other specific applications.

Carl Douglass (13:26):

It’s really interesting to hear that… I didn’t realize that Victrex has a manufacturing segment of the business, both in the composites and sounds like additive space. That’s good to know. Maybe we should form a support group on this whole PEEK topic.

Robert McKay (13:47):

Not to mislead you, we have traditional manufacturing and additive is still a journey for us just as it’s with you. So we have some of the same challenges in our own operations, so how to make additive manufacturing relevant, how to qualify it, how to make it perform the way it needs to. Everything we’re going to talk about today, we have the same challenges and same journey ahead of us.

Carl Douglass (14:08):

That’s a great transition. I imagine that with that practical manufacturing experience, that had to have been a component that led your efforts to develop PAEK for additive manufacturing, at least in part. Does that sound right?

Robert McKay (14:25):

Yeah. One of the areas that we deal in is… Both in aerospace. You hit the two big ones, aerospace and medical, both very interested in additive manufacturing. They have a wide range of parts from those produced in high series parts to those produced in ones and twos. And then the aerospace side you have the whole maintenance and repair aspect and obsolete tools and these aircraft that have been out there for decades and then having to keep them up. So we definitely had that pull coming in from aerospace. And we also had the customer interest in medical. A lot of the surgeries that we have done are… You get better outcomes if they’re customized. We’re all individuals and we all have different bodies and different bone structures, and to be able to customize implants for each person is hugely valuable. So we had those trends pulling on us, but yet we saw how hard PEEK is to print and to take advantage of all the advantages of additive manufacturing to try to do those things. And that’s what pulled us into this, is to try to make it easier.

Carl Douglass (15:35):

We’ve got a couple unique part geometries on the table here that we’ll talk about in a bit. Those were not printed with PEEK, but I won’t go any further than that. Could you talk a little bit about the familial relationship between PEEK and PAEK and the driving development factors that went into PAEK and the advantages of PAEK materials?

Robert McKay (16:04):

Yeah, there’s a bunch of questions in there, Carl. Yeah, maybe I’ll start with the background on polyaryletherketone and PEEK because there is some misconceptions and we’re guilty of creating some of those misconceptions. So polyaryletherketone is actually the family name. So PEEK is a polyaryletherketone as are… There’s polyketones out there, there’s PEKKs out there. Those of us who dabble and add it to manufacturer, high performance polymers are probably familiar with PEKKs as well. Those are all polyaryletherketones. Where we started out, and we’re going to talk about this LMPAEK product that’s Victrex AM200 filament that you’ve been experimenting with, when we started out with that product, we started in the composite space first, which is also a layer wise process, has a lot of the same challenges that additive manufacturing, and that’s really where we first invented that specific flavor of PEEK.

(17:08):

And at the time we were trying to be smart about differentiation and trying to separate ourselves from the PEEKs that are out in the marketplace. There’s lots of different PEEKs out in the marketplace that’ve been out there for 40 years. And we wanted to make it very apparent that this was something new and exciting and different than what was already in the marketplace. So we messaged it as LMPAEK and that’s the brand. And to some regard, we created a separation between PEEK and LMPAEK that was maybe a little stronger than it had to be. So if you actually look at the chemistry behind LMPAEK or the AM200 product that you’ve been working with, it’s really close to PEEK. Essentially, has the same exact chemical building blocks. It has them in largely the same ratios. If you look at the polymer backbone, the way these things are all arranged, 80% of it is exactly the same as you would get in a injection molding PEEK. And the only difference is a slight rearrangement, which changes everything when it comes down to manufacturing.

(18:15):

So with LMPAEK you’re dealing with something that is extremely close to PEEK that people are used to in injection molding world and that we’ve been making for 40 years, and we can get confused by the polyaryletherketone kind of communication. That’s the whole family, is what that is. Silvia, do you want to just elaborate on some of the why we did this thing?

Silvia Berretta (18:44):

Yeah, I think you said this at the beginning as well. So printing with PEEK, it’s possible, but it’s a challenge and nearly 100% of the time it requires an end user that can really trigger the profile settings for a specific geometry. So you have some challenges. It’s not plug and play. Well, first you have to have a nozzle and a machine that goes to the right temperature, but it’s not a plug and play. So when we looked at LMPAEK, we tried to change… At molecular level, we tried to change really some characteristics. So LMPAEK has a 40 degrees lower temperature than PEEK, for example, melting temperature, but has the same glass transition temperature than PEEK. This means that you can shift a little bit your nozzle temperature, but in terms of performance on the parts, you will have very similar performance of PEEK for the same range of temperatures. This is done not by adding additives, it’s done by designing the [inaudible 00:19:50] slightly different.

(19:54):

By doing these changes, we changed as well the crystallization speed of the polymers. So in general, just to make everyone aware, polymers can be amorphous or they can be crystalline. Amorphous polymers can be polycarbonate, polyetherimide, some amorphous nylon, and then you have crystalline polymers like certain type of nylons and PAEKs. [inaudible 00:20:20] fusion having an amorphous polymer might help. And crystalline polymer might be a little bit difficult. Why? Because when they go… When an amorphous polymer get deposited, it will not crystallize, it will stay there. Instead, with a crystalline polymer, it melts, it cools down when it’s deposited above layers, layer upon layer, and then it might moves because it starts to arrange itself, crystallize, and maybe warp or in very extreme case, complete delamination of your part.

(20:54):

So with LMPAEK we change the crystallization speed, so it crystallizes much more slowly than PEEK. So this allows to have to basically widen your processing windows in filament fusion. The polymer will not crystallize as fast, it will be able to be deposited a bit easier. The part will not move, will not distort. But also, this characteristics allows it to have a better diffusion of the polymer chains between layers, which ultimately it will impact the interlayer addition and ultimately the strength of your parts. LMPAEK has also another characteristics, so I don’t know if you have a look at the TDS of PEEK material and you’ll think, “Oh, that has a very good viscosity.” Problem is that viscosity is measured in a shear range that represent injection molding. So a situation where the polymer really has applied lots of pressure to flow into a mold.

(21:58):

That’s not the case in filament infusion. The filament goes through a nozzle and then it flows in open air. So there’s nearly no force applied. And for viscosity, that matters. So for LMPAEK with this material, we have a much lower viscosity in a shear range or non-shear range that represent filament fusion and it’s much lower than PEEK. So all these characteristics help to create a polymer that will have better interlayer strength, a wider processing window that can help you also to build maybe bigger parts and more articulated geometries. And then it’s a bit more able to use at the first attempt.

Carl Douglass (22:42):

We’ve certainly experienced that in practice and it’s helpful to understand the chemistry and the mechanics and the polymer science behind what we get to experience at the printer level. As you know, it has a significant impact.

Silvia Berretta (23:00):

Yeah. And also the fact that you can print a lower temperature enables you to use, for example, soluble supports that are available on the market. So again, even more articulated geometries that may be with PEEK, that could have been very challenging or very expensive because maybe you would’ve used PEEK as well as a support material, which is costly, really. Yeah, and we also AM200 can easily be printed amorphous or crystalline. So it could be printed directly crystalline or it could be printed amorphous, and then going through a post-processing annealing process in order to get all the strength and the chemical resistance and the thermal stability. We have some feedback that is AM LMPAEK 200 is a bit easier to anneal than, for example, PEKK because it seems to be in a sweet spot in terms of crystallization speed. So it doesn’t deform as much as a PEKK would in a non-restrained air environment annealing process.

Robert McKay (24:14):

Yeah, I think as you were saying earlier, sometimes you can always find workarounds if you have enough time. There are already in the market, slower crystallizing polyaryletherketones like the amorphous-tending PEKKs. Those do print better. I think a lot of… You may have even done some of your own work with amorphous poly… PEKKs. The challenge with them is that they, as Silvia was saying, they do have a different crystallization behavior, so it’s hard to get them crystalline in their final form. So a lot of times you’re stuck with them amorphous, which gives up some of the chemical resistance.

(24:55):

The other thing that they bring is a little bit more uncertainty because they are different. So the E and the K’s mean something in these little acronyms. They’re essentially different chemical structures. There’s what’s called ether [inaudible 00:25:12] and ketone linkages that build these polyaryletherketones. So PEEK has two ether and one ketone, PEKK has two ketones and one ether, which brings some different performance properties. LMPAEK has still two ethers and one ketone just like PEEK. So it’s much more similar to PEEK chemistry than some of the alternatives.

(25:36):

And then you have… We see customers try to take polyetherimides or the SOBIC trade name Ultem and make that work because it is a higher temperature polymer, but that is completely different chemistry with a very different behavior. It does work in some cases, but you have to do all the engineering to prove that it works and be comfortable with that to be able to make that kind of change.

Carl Douglass (25:58):

I’ve got one takeaway. There’s a difference between the E’s and the K’s.

Robert McKay (26:01):

And for us MEs, that’s about as far as where we’re going to get into chemistry, right?

Carl Douglass (26:01):

Exactly.

Robert McKay (26:14):

Which is more than maybe some of our audience wants to.

Carl Douglass (26:18):

Well, I just wanted to show my true colors.

(26:24):

So Victrex has been developing and supplying LMPAEK filament for FDM applications for some time. However, we became connected with Victrex through the Stratasys OpenAM and Validated Materials program. Could you talk just a little bit about… That’s a new venture for Victrex, that’s a new venture for Stratasys. What does that look like and how do you see that impacting the market?

Robert McKay (27:00):

We’re tremendously excited by it because of the acumen of the Stratasys ecosystem and the already existing machine base that’s already out there of customers and clients that know how to run those machines and use them for production applications. Up until this time, and it’s, as you said, new for both Stratasys and for Victrex, the Stratasys ecosystem was closed. So they made some and offered some of their own materials, but there was a limited number of them. We’ve known for some time that LMPAEK and Victrex AM200, which is the filament version of LMPAEK, would run on Stratasys equipment. But anyone owning a Stratasys Fortus 450 or F900 couldn’t access it. So they weren’t able to get a spool and throw it on the machine and see how it runs.

(27:49):

So it’s really exciting what Stratasys is doing to bring more materials to their customers and more material innovation and we feel very privileged to be in the first wave of that activity. There’s lots of challenges to that. Stratasys has… You’ve got very specific specifications to be able to run well in their machines that we have to be able to meet. We’re working together [inaudible 00:28:18] over the last… It’s been over a couple of years now, I think, we’ve been working together. But we’re really excited to be at the other end of that pipe.

(28:24):

And I think you were one of the first beta customers to get ahold of some AM200 spools and it’s so exciting when that happens. So I think we all secretly emailed around, “DI lab, just got some spools.”

Silvia Berretta (28:38):

That’s true.

Robert McKay (28:38):

It’s a moment. It’s a really nice moment to see how things work when they get in a real user’s hands who’s going to put them on a machine and try to make things. So really happy that that’s happened and happy to continue to work together with you and Stratasys and your customers even as this continues.

Carl Douglass (28:55):

We are too. When we heard… So at that time we didn’t have a 450. This was a couple years ago now when we were talking to Stratasys and we were exploring expansion of our high performance polymer extrusion side of the business and we’ve always been intrigued by their Stratasys machines, but we’ve always run opensource because that’s what we have to do obviously for these other materials. And they shared with us that they were going down this path and that they were open to having us be the first beta. And PEEK was, in this case it was PAEK but we didn’t know it at the time, was the driver behind that. So we’ve been really looking forward to getting to this stage. For us, in talking to Stratasys has been a couple years in the making.

(29:45):

So when those boxes arrived of the filament, arrived in our office, we were also very excited. I think it was physical interaction rather than… Going and talking to each other rather than email, but we’re in a smaller business. Right away we got to work with the LMPAEK and running it on our machine and produced… This might be the first part that we produced. So this really unique organic shaped geometry was the first part that we produced. And it was really a test. We loaded the LM 200 along with the soluble support material into the machine. We set the profile, just the base profile and went to town and this part came out the very first time we ran it and we were all blown away. In fact, we went into that thinking that, “We’re going to have to do a lot of work in order to get this part to produce just because of the geometry and wall thicknesses.”

(30:57):

With us, you went from us being extraordinarily excited to see the boxes on the shipping and receiving area to now we’re your biggest fans because the amount of time that it would’ve taken us to even approach a geometry like this with an alternative material would’ve been hundreds of hours. We see this as a transformational material that’s going to enable applications that simply weren’t possible before. Not because they couldn’t be printed necessarily, but because just the economics and the time investment just wasn’t possible. So we’re really excited about the application opportunities with LMPAEK, see it as simply transformational.

Brian Douglass (31:44):

We were excited to get those boxes in, but I think there was also a hint of stress and anxiety not knowing what the outcome would be. From our past experience as Carl shared, these high performance materials, they never operate right out of the box. It requires a ton of learning and process development so that we can apply that to geometries. And for us with LMPAEK, it’s been a reverse. It’s been mind-blowing for us to plug in really organic geometries and have those produced where we’re fine-tuning some of the parameters, but we’re not starting from scratch. It’s like we’re making minor tweaks to it that have a pretty big outcome, but the geometries can be produced relatively easy with LMPAEK and this is really exciting for us.

Robert McKay (32:41):

Yeah, we know it’s not, it’s really wonderful to hear that feedback because that was the goal with the material, but we also know there’s still a lot of work ahead and you mentioned some of those minor tweaks you have to make to be able to tune it. It was really nice to be able to see someone who knew how to tackle that in DI Labs. I know you had run some PEEK before and had already gotten a feel for what knobs to turn to try to optimize it and how quickly you made some tweaks and got even better performance out than you got right out of the gate. There’s still learning ahead of us. It’s certainly, maybe compared to PEEK, and it’s kind of a old traditional injection molding form. It’s almost plug and play. It’s certainly not plug and play all the way. Though I’m hopeful as more of us get to work with it and do some more optimization and Stratasys continues their work on it, on their machines, we learn more about how to advise customers to set profiles that would get even easier.

Brian Douglass (33:37):

Yeah, we’re probably in a unique position with our experience with PEAK and [inaudible 00:33:42] and the other P’s out there that we’ve had to produce parts out of. We’ve really skinned our knees and built a great knowledge base so that we can operate within some of these areas and be really effective at iterating process and coming out with an effective printed part. We say it was easy, but it may not be… Yeah, you’re right. It may not be the same case for everyone.

Robert McKay (34:09):

Certainly is in PEEK.

Brian Douglass (34:11):

Yeah, that’s for sure.

Carl Douglass (34:14):

So you were talking, it was really interesting to hear your story about the introduction of LMPAEK, Robert, and the desire to create differentiation between PAEK and PEEK, and maybe even to the extent that there was too much differentiation created because now we’re in a place where we want them to be similar because we want PAEK to be swapped out for PEEK or vice versa. Can you talk to us a little bit about what you’re seeing in terms of acceptance of PAEK in traditional PEEK applications? What are the successes of adopting PAEK in place of PEEK?

Robert McKay (34:53):

So I’ll start and then Silvia will also share some of the recent work we’ve been doing. First just to recognize that we… As you mentioned earlier, there’s two flagship markets of greatest interest, aerospace and medical, are also tend to be some of the more regulated and more conservative industries. So there are certainly going to be early adopters and there are also others that are going to take more time to feel fully comfortable with… Whether or not it’s very similar or not, it still performs different. And in the end, for qualifications in aerospace and medical, it’s not the material alone, it’s the manufacturing process plus the material that has to qualify. So even if you had PEEK, you’d still have to qualify in additive manufacturing PEEK in aerospace and medical, and that will take you just as much work and actually be harder because it doesn’t print as easily. But you still have to do that work.

(35:53):

I don’t want to mislead us and think, “Oh, it’s just plug and play, drop it in and it’s airspace qualified today we can start making flight components.” That’s not the case. We have a long journey ahead of us, but we are pretty comfortable that this material is going to, as we go through that journey together with a machine material pairing like the Stratasys combination with Victrex AM200 filament, that we will successfully get to that end goal. And one example in composites, we mentioned composites as the first adopter of this LMPAEK material. It’s already being qualified in composite form for flight uses. I think it’s Wichita State University, their NIAR center who does some of the aerospace qualifications, has already qualified some LMPAEK thermoplastic composite materials that include this polymer, exact same polymer, that have qualified for flight uses.

(36:52):

Those of us, your customers who are in the aerospace world who may have dabbled in composites, we call that Victrex AE 250 composites is the trade name in the composites world. This polymer that you’re printing with is the same one that’s in Victrex AE 250. Viscosity’s tuned for the additive process, but the polymer chemistry is the same. So it’s already being qualified in aerospace. I think some of the lower or the quick adopters are the places where you need this performance, you need… Some of those parts you showed there where you have a complex manifold or fluid engineering, the industrial, you’re trying to solve a machine problem in a manufacturing environment where you don’t have to qualify for flight or medical implant, their adoption can happen quickly.

(37:43):

But we will have to go through a journey on aerospace, commercial flight especially. You’ll see maybe early adoption in some space, satellite, defense applications perhaps, some low sensitivity applications, interiors perhaps. But the further you get into higher performance flight components and the closer you to get to medical implants, it’s going to be a lot longer journey. Silvia’s been doing some work to help us close some of that gap. Maybe you could share some of that, Silvia?

Silvia Berretta (38:10):

Yeah, I think one of the main things we need to tackle with LMPARK is that PEEK has been around for 40 years, and so it has been used for 40 years and we had nearly 40 years to go after various certifications and qualifications. So one of the first things that I’m looking at as part of my role is to identify what certification we need to go for first and what are the most important for our customers. We already done some of them and the results are very good. They’re very positive because they’re very similar to PEEK. I’m talking about UL flammability or [inaudible 00:38:54], initial bio-cytotoxicity for example. And the results are very positive, very similar to PEEK exactly past in the same range.

(39:06):

We also know that one of the applications where PEEK is used is in the corrosive environment. So we are looking at the chemical resistance of LMPAEK. And we are doing this in a deeper level. So we are looking, okay, what if it’s molded? What if it is left amorph, is printed but left amorphous, and what if instead is printed and crystalline? And then we could see that of course, as expected, we can see that amorphous are much less resistant to chemicals than crystalline forms.

Robert McKay (39:43):

Yeah, you can choose to leave it amorphous like this or you could crystallize it like that. And if you crystallize it, you’re certainly going to get higher chemical resistance. If you leave it amorphous, we certainly see that the chemical resistance printed and even molded amorphous is lower than you would have with crystalline PEEK, with an amorphous LMPAEK. But when you crystallize it all the way, which you can do with LMPAEK, then your chemical resistance is really right in the same range as PEEK. Everything that Silvia’s tested so far has pretty much come up identical as PEEK.

(40:22):

Now, certainly we have… There’s a huge world out there of [inaudible 00:40:26] chemicals that we haven’t tested them all that. We still have a journey to go on there, but the early work we’ve done is showing a pretty good performance in chemical resistance and corrosion resistance of this LMPAEK versus your traditional injection molding PEEK. As long as you crystallize it. Now, not every application needs to be crystallized. You still get pretty good chemical resistance, even amorphous, and that’s one of the pretty cool things with additive is you can choose. Do I leave it amorphous or do I crystallize it? Do I do my prototyping amorphous and then do the annealing afterwards once I’m comfortable with the shape and the function of part? You have that choice. In molding, you don’t get that choice. It comes out crystalline.

Brian Douglass (41:06):

It’s really powerful to see the color difference in the two parts that you have there. The one that’s morphous, the one that’s crystalline. I don’t think we’ve seen that drastic of a color difference on any of the documentation that I’ve gone through.

Carl Douglass (41:24):

No, I’d say most of the parts that we’re producing are semi-crystalline, so they’re between fully amorphous and fully crystalline.

Robert McKay (41:32):

You’re on the Stratasys equipment and your chamber temperatures are probably up there. I don’t know what your settings are, like 130, 140, something in that range? You can really quench it harder. So if you wanted to make it more amorphous, you could. I wouldn’t, but you could if you wanted to.

Carl Douglass (41:49):

So what temperatures? Do you know, off the top of your head, Robert, what temperatures we’d want to anneal these at, or temperatures or times, and times to get to a fully crystalline state? I’m sure it’s dependent on the thickness and everything.

Robert McKay (42:04):

Yeah, I’m going to leave that for Silvia. She’s been doing a lot of work in that area. Usually, it’s in the 161, 180 range for some hours, but depending on if you’re already starting with some crystallinity, it can be faster. And you can always ramp faster or ramp slower depending on the situation. The one thing we do see, and maybe you’re doing this in your work, some geometries are less stiff or resistant to deformation then others. So certainly you can open air anneal this more than you could some alternatives, but in some sensitive geometries, anything really with very low geometric stiffness, you will see it collapse on itself or deform as you try to anneal it. And so an annealing in sand or other kind of inert media like that can solve that problem and eliminate it for large part.

Silvia Berretta (42:57):

We would recommend it to stay away from… So in order to anneal, you need to be above the gas transition temperature of the polymer, which is a 150 degrees Celsius. So we say at least 10 degrees, so 160s or above. Of course, it will need to be maybe… We will never recommend it to put your oven or furnace at temperature and then throw it your parts, it give the part a bit of a thermal shock. We will recommend it to slowly heat the part up to temperature within the oven and leave it [inaudible 00:43:38] in temperature of choice, the one that you maybe you optimize for a particular geometry, for a couple of hours at least. And then it might need some tweaking. And then slowly cool down as well.

(43:50):

We know that it’s possible and it helps to maintain geometry to anneal in different medias. So sand or gypsum. It could vary a bit on your part. We also know there’s a rule of thumb in general with the PAEK materials, is that if you know that your part will be used at a certain temperature, so let’s call it your applications or service temperature, then anneal 20 degrees higher than that because that will ensure that when the part is in place in the application, it will not change, it will not go into any relaxation.

Carl Douglass (44:34):

That’s really helpful. We were talking about the fact that as we’re producing these parts, we’re probably producing them in a semi crystalline state, so not fully amorphous and also not fully crystalline. And it was really helpful to see… The Robert shared a couple of parts on the screen. It was really helpful to see the stark contrast in color of those two parts that allows us to push the boundaries a little bit more on getting the parts fully crystalline through the post-processing.

Silvia Berretta (45:06):

Yeah, and it’s good when you see only one color on the part. It becomes a problem where you see both and it was not desired. Yeah.

Carl Douglass (45:16):

Do you foresee any time where Victrex might start to introduce reinforcement additives into the filament, like carbon fiber or fiberglass?

Silvia Berretta (45:29):

Definitely, yes. Yeah, it’s something we are already working on in the background. So yes, definitely.

Carl Douglass (45:37):

That’s something that we’ll be mighty excited to… We’re awaiting that now as we speak and we’ll be excited for that launch. In fact, if you’re looking for any beta users of that, we’d be happy to do that. There’s a lot of applications that we venture into that require that level of stiffness or reinforcement. From our experiences so far, being able to work with LMPAEK as a base material would be as transformational as what we’ve already experienced. There’s no doubt about it.

Silvia Berretta (46:12):

And just be mindful that according to the fillers you might need to have some additional checks. So just to be, [inaudible 00:46:22]. For carbon fiber parts, for example, they will be black, so you will not be able to see if your print is stable, your printing process is fully stabilized because you will not see what is amorphous and what’s crystalline. It’s all black. So then when a filament like this is in your process, you need to make sure as well to check really the crystallinity content in your parts and to be methodical and how will be… It’s part of the journey with fillers really, especially with black fillers. Yes.

Robert McKay (46:59):

Yeah, we’ve had a lot of requests for black filament because it hides the crystallinity but there’s a double-edged sword to that. If you hide the crystallinity, you can’t see how in control your process is and you want to be able to see it if you can. Now, black carbon filler, you won’t be able to. So as Silvia has said, you’ll have to overcome that different ways. Yeah, it’s something to talk about. So we have a prototype, and we talked about this last year at CAMX, at the CAMX conference. Silvia presented there. If you were part of the proceedings, you can go look that presentation up. We can send you the content from that as well. We are doing a short carbon-filled fiber version of AM200 filament. That’s in the plan and there’s already prototypes available. So with your full open material license that you have access to, we may even be able to get material to you now if we talk that through with Stratasys. Rather than wait for Stratasys to evaluate it and qualify it, we could start that now.

Silvia Berretta (48:06):

Do you see any particular requests for some type of [inaudible 00:48:10]?

Carl Douglass (48:10):

I’d say the fiber filled is probably the most prevalent, whether it’s carbon or glass fiber.

Brian Douglass (48:20):

If we can run glass fiber over carbon fiber, that would be our preference just because of the reduction in abrasion that we see on the equipment. We lose some stiffness there through that glass fiber reinforced rather than carbon fiber. But if the demand isn’t there on the application, then I think there’s an advantage on the processing side.

Robert McKay (48:44):

We just ran some glass fiber prototypes a couple weeks ago, so the work’s being done I think… And that is a great question Silvia asked because we get the maybe misleading impression that carbon fiber is the more in demand. We’re starting with carbon fiber, but if it’s the other way around, we can always change the schedule.

Brian Douglass (49:04):

That’d be interesting to review some of our recent projects where we’ve had specific stiffness requirements for geometries we’ve produced. And then if you’ve got material properties off of your finished carbon fiber versus glass fiber, then we can compare those and see where that would end up.

Carl Douglass (49:26):

In the AM industry, it’s still quite young. Oftentimes we’re approaching projects or we find that projects are approached in a little bit cavalierly. And not necessarily in a negative way, but we don’t have the legacy of injection-molded materials with injection-molded material testing to really understand what provides us with the greatest strength or suitable characteristics for our applications. So there’s a lot of cases where we’ll find that parts may be over-specified based off of intuitive or just practical knowledge rather than truly understanding the needs. And that’s not a bad thing because in some cases it’s okay, in other cases it’s just overkill.

(50:16):

And there’s this other component that the whole conversation of crystallinity versus amorphous, those aspects of the conversations usually don’t take form in specification. And so there’s this side of additive that needs to mature yet so that we really understand what we’re specifying to make sure that it fulfills the application. And in the case of carbon fiber requests, I’d guess that there’s a lot of carbon fiber requests because there’s some conservatism in that over fiberglass. Obviously, that’s not talking to all applications, it’s just an anecdotal perspective. But we see that quite often in our scenario.

Robert McKay (51:01):

Something we’ve seen it is the reverse where it’s the print bureaus themselves that are driving carbon fiber because they get a customer coming in and says, “I want PEEK.” And then they try to run PEEK like you described, and they have a miserable time of it. And then, “If I put a little filler in there, it starts to counteract [inaudible 00:51:21] shrink and warp. It makes it a little easier to work with.” So I can go back to these customers and say, “Use this carbon fiber PEEK because I can actually print it when I couldn’t print unfilled PEEK.” We feel like some of that will go away with LMPAEK and Victrex AM200 filament where you can use the unfilled PEAK and get it to print well enough and not have to throw the fillers in there to solve the problem. Yet there are still.

(51:46):

So that’s a piece. There are still applications out there that want filler for good reasons, temperature resistance. So these polymers soften as they get past their softening temperature, the Tg. And if you have fillers in there, you can boost that 10, 20. Your use temperature 10, 20, 30 degrees depending on the situation. So that’s a real case for having filled filaments. And so we know we’re going to need filled filaments. But in the cases where you can get AM200 to do what you need it to do, you don’t have to throw the filler in there just to fix the problem.

Carl Douglass (52:22):

Agreed. This has been a really enlightening discussion. Outstanding to hear the history of PEEK and PEAK, and the adoption and testing that’s being done on PAEK to drive greater adoption. If you were to give advice to an organization that has traditionally used PEEK and could explore PEAK, what advice would you give them?

Silvia Berretta (52:53):

Maybe from my side, I would say if you can pick up all the failures that you had with PEEK, have a go with LMPAEK and see what happens. And what we’d expect is similar to what you said really that geometries that you [inaudible 00:53:14]. Yeah, they will be printable with PAEK, but after how much time. You will see them printed out the first time right. So I think that would be… It’s such a significant difference. That would be my advice. Pick up all the failures and try them with LMPAEK.

Robert McKay (53:34):

And mine’s similar. You’re going to have to go through a journey. Even if you had PEEK, you still have to qualify on the machine and the printed part. And you’ve shared some of the challenges that most of our customers actually ignore that, “Give us PEEK.” We tell them, “You’re going to have a hard time of it,” and they insist on having it anyways. And they go and try to print it and then they come back and say, “It didn’t work very well.” So my advice would be, understand you have a new process here. You’re going to have to go through qualification process. Why not go through a qualification process you have a chance of succeeding in? And the closest material that actually works really well is LMPAEK.

(54:09):

And so choose that to go through that journey rather than trying to force the other things that are maybe a lot different than PEEK and trying to do it that way or trying to force PEEK to work when it really doesn’t work very well. And maybe that’s a bit of an overconfident statement in terms of choose LMPAEK as your candidate to go through the process with, but that’s how it feels for us right now.

Carl Douglass (54:32):

It’s well-founded from my perspective with the history that you both have and the history that Victrex has in polymer development in both PEEK and PAEK, and the manufacturing operations that you have. To me, that resonates very clearly that there’s a strong foundation that’s driving that feedback and advice. So thank you both for that.

Robert McKay (55:00):

Thank you for the invitation.

Silvia Berretta (55:00):

Thank you for inviting us.

Brian Douglass (55:03):

My takeaway is that as we first heard about LMPAEK and not digging into the details, our sights were on PEEK and we had to have PEEK because that’s what was demanded. And we first heard about LMPEK, we were like, “This isn’t what we need. We need PEEK. We need to be able to specify this to clients.” But after this discussion, I think that we’ve got some confidence and foundation built that we can better align LMPEL with PEEK applications because the differences are small. And we’ll work on communicating that through rigid data that we can provide, but we really we’re not as far off as what we expected we would be from PEEK to LMPAEK. So I think our objective coming out of this is working on our communication and terminology so that we can tie those together as we’re working on applications.

Robert McKay (55:55):

I hope you bring those parts next week to wrap it. I can’t see them up and close, but from this distance, you should be proud of those. Those are really good-looking parts from really complex parts. And to get them out of your machine with only… Obviously, you had your expertise and doing some the tweaks you did, but you should be proud of that output. I can’t wait to see them up close.

Carl Douglass (56:15):

Now we have to bring him, Robert.

Robert McKay (56:16):

Yeah, you do. Another suitcase.

Carl Douglass (56:23):

Thanks so much, Robert. This has been a lot of fun, very informative and absolutely one of the most enjoyable topics that we’ve got. We’re looking forward to helping expand the adoption in the market for LMPEK because we see that this really starts to open the doors to applications that otherwise wouldn’t be easily solved.

Robert McKay (56:49):

Thanks, Carl.

Carl Douglass (56:50):

Thank you both.

Silvia Berretta (56:52):

Thank you.

Robert McKay (56:53):

[inaudible 00:56:53].

Silvia Berretta (56:58):

Thank you.