AS: Hi everybody. Welcome to today’s Power Corner. I’m your host, Aalyia Shaukat, Editor-in-Chief, Power Electronics News. Today, we have the pleasure of speaking with Guido Überreiter, VP of Semiconductor Strategy at Von Ardenne. Guido, how are you doing today?
GÜ: Very good. Thank you. Thank you for the opportunity to talk here.
AS: Thank you for joining me. I really appreciate it. So I want to do just a quick introduction. I did a little bit of research about Von Ardenne. From what I understand, it’s a German vacuum-coating equipment company that, despite having 1,000-plus systems operating in 50-plus countries, is still a family business run by Pia Von Ardenne. And from what I understand, within the semiconductor space, Von Ardenne focuses specifically on physical vapor deposition, so PVD.
The company makes deposition equipment that goes into MEMS, piezo-MEMS, and advanced packaging, which, for example, integrates power semiconductors with logic chips. And I understand that Von Ardenne researchers are actively developing thin-film gallium oxide (Ga₂O₃) growth methods tailored for device fabrication.
So maybe we can first start off with Ga₂O₃ as a semiconductor. What are its benefits, and what makes it a compelling candidate for high-voltage power devices beyond 1 kV?
GÜ: Yeah, I think one of the interesting aspects of wide bandgap is that you are actually looking for a wider and wider bandgap to be able to switch higher and higher voltages. So we have silicon carbide (SiC) today, and we have gallium nitride (GaN) today.
SiC actually works in a similar voltage range, but there are multiple benefits of Ga₂O₃ that we can talk about later. But the industry is looking for better and better power semiconductors, simply driven by, I think, the growing importance of power conversion and efficient power conversion. Initially, we talked about SiC and GaN a lot in the context of the electrification of vehicles. And over the last two, three years, with the AI explosion, suddenly everybody is also looking at the megawatt data center kind of contributions.
And if you have in the back of your mind that 20–25% of the energy is lost just going from power generation through all of the distribution until it reaches the actual CPU, there’s a lot of energy-saving potential if we make better and more efficient power semiconductors.
And Ga₂O₃ is one way of doing it. As every material comes with a lot of benefits, it also comes with a couple of downsides that we can talk about. But overall, I think it’s a very interesting material that is still in a bit of an early stage, but it’s promising significant advances from an efficiency and cost perspective.
AS: Yeah. So, as you were saying, Ga₂O₃ has been studied for decades. What changed in deposition technology that makes it viable for power device manufacturing today?
GÜ: Yeah, I think it’s sometimes this demand-and-supply kind of dilemma, right? If you’re too early and there is not enough demand, and people are happy with what they have, it’s not picking up.
But I think there are multiple things that have changed. One is that there is a lot of pull, as I said. The second is, I think we have now developed methods to create Ga₂O₃ that are much more reliable and much more proficient at generating the material in a way that it’s actually relevant for making power semiconductors out of it.
And on top of that, I think the thin-film technologies in general have now progressed enough that you can also make very detailed and very fine layers on top of the actual substrate—because it’s not just the material itself, it’s how you are actually using it. And that’s where I think Von Ardenne in particular is coming from: a lot of different materials and a lot of different ways of depositing them.
That’s where we see—the time is ready for utilizing PVD more on those kinds of applications as well. And I think, as I said, there is a lot of pull—EVs, grid conversion—all of those things are coming together. And I think it’s also the point where there could be another small revolution in power semiconductors.
AS: Maybe we could talk a little bit about that substrate growth technology that Von Ardenne is working on, and describe the advantage of Ga₂O₃ versus SiC or GaN in terms of substrate growth scalability.
GÜ: It’s actually—when I joined Von Ardenne just a couple of years ago, I was always amazed by how they use the materials. They are using certain materials already for years that just now make their way into semiconductors—PVD is, in general, a very cost-efficient way of depositing materials.
And that’s actually one of the downsides of SiC and GaN: making those wafers is still very, very expensive. It’s relatively hard to make them on large substrates. So, going into 300 mm—the industry has been fighting very, very hard to get SiC onto 300-mm, we’re slowly getting there on all sides of the globe. But still, it’s a very, very expensive approach, very time-consuming, and also very energy-consuming.
And that’s where I think PVD technologies can make a difference—not only in adding layers on top of a substrate, but one of the technologies we are currently developing is actually growing the material in the specific crystallographic structure with PVD technologies. So it’s not just putting layers on top; it’s actually also building the substrate itself in a PVD flow, which is kind of novel, and it’s even more cost-effective because you can actually stay in the same toolset.
AS: So, just as an example—let me know if I’m missing something—I know that GaN needs to be grown on specific Si [111] wafers. Is it kind of like that, or, as you were saying, that’s not necessary? Can you match the structure as you grow Ga₂O₃?
GÜ: Our technology is able to basically create particular crystallographic structures atom layer by atom layer. So we have a tool that is actually able to grow, and you can tune the process setup in a way that you can get the physical parameters—and with that also the electrical parameters—that you want to have.
GaN, when you look at GaN structures, in many cases it’s not just GaN. It’s GaN, and then it’s a little bit of doping here, and then it’s another GaN layer, and then it’s gallium doped with aluminum.
So there are a lot of mixed-layer stacks. The fundamental question is always: how do you build the base layer? And we believe we can do both.
Whereas in GaN you basically need MOCVD-type processes to build the actual substrates, that’s where we believe we actually have an edge—utilizing an element we have worked with for a long time already, and deploying a couple of tricks on the PVD side to basically generate the base layer, but also the layer stacks on top that are required because of some of the deficiencies that Ga₂O₃ has as a material.
AS: Okay. And maybe we could talk a little bit more about Ga₂O₃’s appeal for commercial adoption. From what I’ve researched, p-type doping has been a long-standing challenge in Ga₂O₃ for full device integration—creating the p-n structures needed for diodes and transistors. Does Von Ardenne’s deposition approach have any role in solving the p-type challenge? Or is that a separate problem that needs to be solved upstream?
GÜ: Well, I think it’s actually material-inherent, not the way you make the material. The fact that you cannot make p-type doping—or that it’s very hard and actually very inefficient and therefore not desirable—is something that we will not change, whatever method we use in making the material.
But what that drives is a lot of innovation on how you actually build devices on top of that layer—either with Schottky-type devices or with MOSFETs. The industry typically then goes into the third dimension, building layers on top and building devices in a different way than just embedding them in the actual Ga₂O₃.
So that’s what we have seen with other approaches as well—GaN is, not from a p-type doping perspective, but from a device architecture perspective, it’s very similar. There are a lot of vertical devices on GaN film, and that’s what we see in Ga₂O₃ as well: overcoming that material challenge with other device architectures.
AS: Interesting. Do you mind talking a little bit more about those alternative device architectures with Ga₂O₃?
GÜ: Well, if you want to build other types of devices, you basically start again by adding layers on top with different materials or differently doped materials. And what’s then super important is the interfaces of those layers.
In GaN’s case, for example, how exactly does GaN and AlGaN interact at the interface, and create a very clean surface that you can very nicely tune the device parameters with?
And that’s what we will see with Ga₂O₃ as well: devices that basically go vertical, where you have those junctions on top of each other instead of left and right of each other. And I think that’s what the industry is adopting more and more, also to optimize on SiC and GaN. So the same methodology can be applied when you go to Ga₂O₃.
What will be important is being able to very precisely create the surface, very precisely deposit the next layers—which will be very thin layers again—and have a clean surface and a clean interface. And with those alternating layer stacks, you can create devices that leverage Ga₂O₃’s wide bandgap and high electric-field capabilities.
AS: Thank you. And we can talk a little bit more about the PVD techniques and the benefits of that. You mentioned magnetron sputtering or pulsed laser deposition. How does that compare to, say, metal-organic chemical vapor deposition (MOCVD) or hydride vapor phase epitaxy (HVPE), in terms of film quality, scalability, and cost?
GÜ: Well, I think in general, PVD is very cost-effective. It’s much cheaper, it’s much faster. And if you think about some of our tools that are basically processing dozens of wafers in parallel as they are spinning in the tool and as we are depositing atomic layer by atomic layer at high speed, you can see how that creates a cost advantage overall for making those wafers.
And as we are not limited by wafer diameter, we can basically go as large as the substrates we can get to grow things on. That’s actually another advantage that plays into cost as well. The interfaces I mentioned—making clean surfaces, making them very smooth and very homogeneous—are something that PVD clearly does better than CVD technologies or similar.
What we have to do—and this is the engineering effort on our side—is to make sure that the crystal structure is exactly as we want it to be. I think that’s where the engineering excellence and the material science come into play on PVD, because that’s what, for a long time, has been pushing applications toward those more expensive deposition technologies.
But we believe we have some very good engineers who know how to mix the concrete, so to speak. And that’s where I think PVD will get a much, much bigger role in power semiconductors because of the progress that engineers have made in producing materials and surfaces.
AS: That’s very interesting. This is personal curiosity—have you created any power devices thus far with the Ga₂O₃ technology that you’ve been working on?
GÜ: No, we’re still in the phase where we are depositing films. We have actually very successfully created what I would call alternating stack films—and not in a way where we move a wafer into the next chamber and back. We basically have a tool that just rotates, and with every rotation, it receives a certain deposition of materials. And as it rotates, you can change the composition of those materials.
So you keep the flywheel running, and as it runs over time, you build first layer A, then layer B, then layer A again, then layer B, and you can mix them very nicely in a very fast way. And that’s where we’re seeing a lot of interest right now—where people say, “Oh, that’s actually how I can make my devices much faster with very different materials.”
AS: That’s very interesting. What application spaces—whatever you can’t reveal, you can’t reveal—are showing the most interest in this?
GÜ: Well, as I said, Ga₂O₃ is the one we are pushing, but the same tools are basically able to do GaN and, you know, mixes of layer stacks, and that’s just the power electronics side.
You can also use it for MEMS applications—to grow particular layer stacks or materials to a very thick stack in a very short time. It’s kind of the same tools, and that’s where we are actually further ahead on the MEMS side.
But basically the same physics and chemistry apply when you go into Ga₂O₃, GaN, or any of those materials. And as you mentioned before, you keep going up in bandgap—there’s boron nitride (BN), and at the end, diamond materials. But it’s basically a similar kind of problem that you need to solve: getting the right crystal structure and getting it very homogeneously grown, very fast.
And we believe, with the tools we have at hand, we can actually switch materials, and with that also switch the electrical and physical parameters of those materials.
AS: Fascinating. Is there anything you wanted to add, or any questions I might have missed? You’ve already answered quite a lot of my questions.
GÜ: Well, I think there’s the cautionary statement, right? Where is Ga₂O₃ from a deployment perspective? Because this is not high-volume production today.
I think we’re still in a phase where a lot of the power semiconductor players are deploying their next-generation SiC factories for high volume. We are seeing a lot of progress in GaN and multiple variants of GaN-based semiconductors.
I think Ga₂O₃ still takes a couple of years to get out of that experimental phase—and this phase where, out of the materials, the application engineers can actually build real devices—into something you would call prototyping or first product.
My personal estimation is that it’s going to take another five years until we see the first products, and then probably another 2-3 years after that before it gets into real high-volume production.
But honestly, I think we all believe we need until 2030 to get to a $1 trillion industry. And now the industry says we are at $1.3 trillion in 2027. So it seems a lot of things are accelerating, and I think that’s also, to some extent, a wake-up call to the traditional power semiconductor houses—because I think the SiC era will be over much sooner than many people think.
AS: Well, thank you so much. I really appreciate your time, Guido. This has been a fascinating discussion. I’m excited to see where Ga₂O₃ goes, and PVD’s role in that, and how Von Ardenne will grow that technology.
GÜ: Perfect. I’m fascinated by it too. Thank you.
AS: Thank you.
The interview was originally published on Power Electronics News.
Click here to watch the interview: PVD's Role for Gallium Oxide Power Semiconductors - Power Electronics News