Welcome to PowerUP, a podcast show hosted by Maurizio Di Paolo Emilio that brings light to some of the stories on power electronics technologies and products featured on PowerElectronicsNews.com and through other AspenCore Media publications. In this show, you’ll hear both engineers and executives discuss news, challenges, and opportunities for power electronics in markets such as automotive, industrial, and consumer. Here is your host, editor-in-chief of PowerElectronicsNews.com and EEWeb.com, Maurizio Di Paolo Emilio.
歡迎收聽由Maurizio Di Paolo Emilio主持的podcast節目Powerup。本節目為 PowerElectronicsNews.com 和其他 AspenCore Media 旗下網站的內容帶來活力。在本節目中，您將聽到工程師和高層討論電力電子和汽車、工業與消費性電子等市場的新聞、挑戰和商機。歡迎我們的主持人，Power Electronics News和EEWeb.com主編Maurizio Di Paolo Emilio。
MAURIZIO DI PAOLO EMILIO: Hello everyone, and welcome to this new episode of PowerUP. Today, we’ll be talking about gallium nitride for high voltages. Wide-bandgap power electronic device advancements have lately made impressive strides, mostly because of today’s higher switching frequencies compared with silicon devices and their consequent capacity to rise the power density and efficiency of switch-mode power converters. Power-switching transistors built with GaN are superior to those built with silicon because the material’s properties enable the realization of devices with a much smaller wafer area and an equivalent breakdown voltage or on-resistance and current-handling capacity.
MAURIZIO DI PAOLO EMILIO： 大家好，歡迎收聽最新一集的PowerUP。今天，我們將討論高壓應用的氮化鎵。寬能隙電力電子元件的進步最近取得了令人矚目的進步，這主要是因為與矽元件相比，它們如今的開關頻率更高，並且因此具有提升開關模式電源轉換器的功率密度和效率的能力。用 GaN 製造的功率開關電晶體優於用矽製造的電晶體，因為該材料的特性能夠實現具有更小晶圓面積和等效擊穿電壓或導通電阻和電流處理能力的元件。
Despite its success to date for lower-voltage applications, about 650 V and under, the most measured gain-based power device, the high-electron–mobility transistor (HEMT) could be unsuitable for high voltage, medium from 1.2-kV to 20-kV applications, including electric-vehicle drivetrains and many great applications, where silicon carbide is going very well. Odyssey Semiconductor has developed a metal to achieve area selective doped regions in GaN, opening the door to the realization of vertical conduction devices, analogs, to the highly developed device geometries that are standard in silicon and SiC but with all the superior material properties of GaN.
儘管迄今為止在約 650 V 及以下的低壓應用中取得了成功，但高電子遷移率電晶體(HEMT) 可能不適用於1.2 kV至20 kV的高壓應用，包括電動車傳動系統和許多出色的應用，其中碳化矽進展順利。 Odyssey Semiconductor開發了一種金屬來實現GaN中的區域選擇性摻雜區，為實現垂直導電元件、類比元件，還有對矽與碳化矽材料材料來說是標準產品、高度發展的元件種類打開了大門，這些元件將具備GaN所有的優異材料特性。
In this podcast with Mark Davidson, CEO of Odyssey Semiconductor, we will discover more about these features. Let’s talk with Mark.
本集Podcast邀請到Odyssey Semiconductor執行長Mark Davidson，我們將更進一步探討這些特性。讓我們和Mark聊聊。
Hi, Mark. Thanks a lot for joining us. How are you?
MARK DAVIDSON: I’m great, Maurizio. Great to be connected with you. Thank you for letting me be involved in this podcast today.
MAURIZIO DI PAOLO EMILIO: Thank you. Thanks a lot for this opportunity. So today, we will talk about GaN for high voltages. But before that, tell us more about you. Please introduce yourself, your company.
MAURIZIO DI PAOLO EMILIO：謝謝你，非常感謝有這個機會。所以，今天我們將討論高壓應用的GaN。但在開始前，請告訴我們更多一點你的資訊，請介紹你自己還有你的公司。
MARK DAVIDSON: Yeah, great. Thanks, Maurizio. So yeah, my name’s Mark Davidson, CEO of Odyssey Semiconductor. I started this position in April of this year. I’ll tell you a little bit about the company and then a little bit about myself. So Odyssey was formed with the focus to create vertical GaN power FETs to go after the high-voltage power market. And throughout this podcast, I’ll give you some more detail as to why and how we’re doing that. But that’s our place in the market. And in summary, I would say our job is to bring the switching performance and power efficiency of GaN into voltages currently being addressed by SiC. That’s kind of our position.
MARK DAVIDSON：好的，很好；謝謝Maurizio。我的名字是Mark Davidson，是Odyssey Semiconductor的執行長，我是從今年4月開始擔任這個職位。我會告訴你一些關於公司的事情，然後是關於我自己的一些事情。Odyssey的成立宗旨是打造垂直GaN功率FET，以高壓電源市場為目標。在這集Podcast裡我會詳細介紹我們為什麼要這麼做，以及會如何做；這是我們的市場定位。總而言之，我想說我們的工作是把GaN的開關性能和電源效率導入目前以SiC元件來因應的電壓應用領域，這是我們的定位。
For myself, [I’ve spent] 20 years in power semiconductors working for a lot of the large established power and analog companies. I started my career as an engineer in automotive doing powertrain electronics design and working with manufacturing. So I have this background in automotive that I’m finding to be useful, as electric vehicles is one of the markets that we’re going to be pursuing. But yeah, 20 years in power semiconductors, and I worked at a bunch of startups as well. And like I said, since April of this year, I have been CEO of Odyssey Semiconductor based out of Ithaca, New York, on the east coast of the U.S.
至於我自己，我在電源半導體領域有20年的經驗，為許多大型老牌電源和類比公司工作過。我的職業生涯一開始是汽車領域的工程師，從事動力系統電子設計和製造。我發現我在汽車產業的背景很有用，因為電動車也是我們的目標市場之一。所以我在電源半導體領域工作了20年，我也在幾家新創公司待過。而就像我前面說的，我從今年4月開始擔任 Odyssey Semiconductor執行長，我們公司總部是在美國東岸的紐約州伊薩卡市
MAURIZIO DI PAOLO EMILIO: So Mark, let’s start with the advantages of GaN compared with silicon, but also SiC, and which are the problems in high voltage for GaN, more than 650 V. So tell us, please explain the differences between lateral and vertical structure.
MAURIZIO DI PAOLO EMILIO：所以Mark，讓我們從GaN相較於矽的優勢開始，還有SiC，以及 GaN在超過650V的高電壓應用問題。所以先請幫我們解釋水平與垂直元件結構的區別。
MARK DAVIDSON: Yeah, great. Yeah, so if we look at power, you know, the three main alternatives that we see are silicon and IGBTs and I think, you know, you’ve done a lot of coverage. We know that silicon is just run out of juice on high voltage, high efficiency, and high power density. So, you know, kind of the movement away from silicon has been happening for a little while. So then we look against lateral GaN and SiC as the two main alternatives to what we’re doing in vertical. So let me start with lateral GaN. And of course, you know, lateral GaN is GaN-on-silicon or GaN-on-sapphire.
MARK DAVIDSON：好的，很好。所以如果我們看一下電源領域，你知道，我們看到的三個主要替代品是矽和 IGBT，我認為，你知道，你們已經做了很多報導。我們知道，矽已經在高電壓、高效率和高功率密度上不敷使用。所以，你知道，尋求矽以外材料的行動已經發生了一段時間。因此，我們將水平式GaN和SiC作為我們在垂直結構上所做的兩個主要替代品。因此，讓我從水平式GaN開始。所謂的水平式GaN是GaN-on-silicon或GaN-on-sapphire。
So we’re talking about really a GaN layer on top of a more traditional substrate. And born from really RF, you know, that the structures are HEMTs and were developed for RF, and now being adapted by a few companies to go after power. And you know, with this lateral device, we see a lot of the lateral GaN companies going after laptop chargers and, you know, trying to move into some of the higher-performance markets. But we see lateral GaN really going into what I would call the mid-range. And one of the challenges, of course, with lateral GaN and ability to go to higher voltages is inherent in the structure. And instead of talking about the theory or the technology, let me give you a real example.
We saw a lateral GaN device, power device from a competitor that was rated at 1,200 V, and we were very curious about that. So we were able to do some competitive teardown on that product, and the one thing I’ll tell you is, their die, for the same on-resistance, and in this space, the die size is dictated by the on-resistance, by the RDS(on). For the same on-resistance, their die is 5× bigger than ours. Because when you go lateral, and if you want to increase the voltage capability of the device, you’ve got to go larger die area, where when you go vertical, you know, you increase the thickness a little bit. So to me, that’s a great example of why lateral GaN is really limited in going after these really high-performance, higher-voltage applications. And then, you know, the last alternative, of course, is SiC. And I mean, look, and you’ve covered it well. SiC is going into the broad adoption phase finally after many years in power.
And that’s great to see, because I appreciate that the SiC companies are making the investment to move the customers from silicon to, you know, a compound semiconductor option. And SiC, you know, obviously it’s significantly better than silicon, but still limited, and still not as good at what vertical GaN’s going to bring. For example, if we look at, and we’re going to talk a little later in the podcast I think about, we’ll call it the figure of merit. But when we look at the capabilities of SiC, first is vertical GaN. Vertical GaN gives about a 6× to 10× improvement over SiC. In the example I gave, lateral versus vertical GaN, I’ll give the same thing. You know, what that 6× to 10× improvement means is, for the same RDS(on) at, let’s say, 1,200 V, you know, six to 10 of my die will fit inside one SiC die.
That’s a significant advantage, not only on induction loss, and of course cost, because cost is proportional to die size, but also on switching frequency, the smaller the device. So GaN is already better than SiC for on switching, and then a smaller die is going to give me an even greater advantage. So we see vertical GaN as that kind of big step. The silicon to SiC step is being taken, and SiC to vertical GaN will be the next step that’s taken, and we’re proud that we’re leading the way in vertical GaN.
MAURIZIO DI PAOLO EMILIO: So your company, Odyssey Semiconductor, is using high-quality bulk GaN wafers as the substrate for your proprietary vertical conduction power-switching transistors. So this vertical conduction device shape is advantageous because it uses the substrate’s surface area effectively and enables the use of a normally off device and, moreover, isolates high voltages that occur between terminals. So in terms of challenges, tell me which are challenges more about your technology that you are working on. Why do we need vertical GaN? You mentioned early in this case, but what are your customers from the design point of view asking you in this case for specific applications?
MAURIZIO DI PAOLO EMILIO：所以你的公司Odyssey Semiconductor正在使用高品質的塊狀GaN晶圓片，作為你們獨家垂直式導電功率開關電晶體的基板。這種垂直式導電元件的形狀是有利的，因為更有效地利用了基板的表面積，並且能夠使用常關元件，此外，它還隔離了端子之間出現的高壓。因此，就挑戰而言，請告訴我哪些是你們正在研究的技術的更多挑戰。為什麼我們需要垂直式GaN？你先前有提到這個案例，從設計的角度來看，你們的客戶是否針對特定應用提出過什麼要求？
MARK DAVIDSON: Yeah, great. So I’ll start with performance, the performance benefits. And actually, we’ll talk about cost of performance. And I talk to a lot of customers regularly. Although I’m CEO, I really come from a marketing and sales background. You know, and one in particular I’ll talk about in automotive OEM that sees vertical GaN as very, very important for electric vehicles. Moving to 800-V batteries, SiC with 1,200-V capability is there now. I think we’ve seen some teardowns of SiC FETs in the traction inverters, but still too expensive, still too large and bulky. So the problem they want to solve is continuing down that price performance curve for 800-V battery packs in a traction inverter. So that’s, you know, I think a great example.
MARK DAVIDSON：好的。我會從性能開始講，性能優勢；實際上，我也會提到性能成本。我經常與很多客戶交談，雖然我是執行長，但我來自行銷和業務背景。我會特別談到的一點是，車廠認為垂直式GaN對電動車非常非常重要，當轉向採用800V電池，具備支援1,200V電壓的SiC已經就緒。我認為我們已經看到牽引逆變器中的SiC FET的一些拆解，但仍然過於昂貴，仍然太大且笨重。因此，他們想要解決的問題是繼續降低牽引逆變器中800V電池組的價格性能曲線。這是我認為一個很好的例子。
And those examples exist in other markets as well, but I think that’s a great example of why the market and customers are looking for vertical GaN. That’s on the cost performance side. If we look on the reliability side, you know, when we go GaN-on-silicon versus GaN-on-GaN, it’s really about defect density. And with our GaN, number one, less defects, and of course, this has a big impact on reliability. But then as we scale, because we scale vertically, we’re not intersecting more defects. So from a reliability point of view, you know, we see vertical GaN-on-GaN as an inherently more reliable solution. Now I’m not saying that the other companies aren’t reliable. They clearly have products that are qualified.
They have found ways to overcome the challenges for long-term reliability. But if you just look at material properties, lower defect count and then less intersection because of vertical scale versus the horizontal scale, I think that’s a really big benefit. And I’ll also remind that when we look at the power, discrete business, power FETs, SiC, silicon, they’re vertical conduction devices, right? Power is really designed to be a vertical conduction transistor. And I think we’re bringing that in GaN where it hasn’t been brought before. So I think, you know, we combine all those things. And yeah, of course we still have challenges to solve. We’re not qualified in production yet.
We’ll talk about it at the end of the podcast, I think, but we have our first samples coming out. They’ll be assembled this year and we’ll put them into customers’ hands next year, and we’re going to learn a lot more. We’re going to learn a lot more about what the customers see, what they observe, what they expect from their products. So, you know, we still have some challenges to solve, but I’ll go back to the inherent capabilities of the material, the capabilities of a vertical structure. You know, we know we’re bringing the right combination of technology and solution into the market.
MAURIZIO DI PAOLO EMILIO: So in terms of manufacturing and cost, so we are talking about vertical GaN. So the manufacturing will be the same obviously on silicon carbide, silicon. So in theory, you are using the same facilities. And in terms of cost, which is the direction? How now and in the future can we compare vertical GaN with the equivalent of silicon and SiC in higher voltages?
MAURIZIO DI PAOLO EMILIO：所以在製造和成本方面，我們談論的是垂直式GaN。因此，無論是在SiC或矽材料上，製造顯然是相同的。所以理論上，使用的是相同的設備。而在成本方面，趨勢走向如何？現在和未來我們如何在更高壓的應用上，比較垂直式GaN與矽和SiC？
MARK DAVIDSON: Yeah. And you know, I’m not going to, if it’s okay with you, I’m not going to talk about the cost comparisons between us and silicon, because for silicon, in the places that we’re going, cost doesn’t matter because they’re not meeting the performance requirements. But I think the big question people have is vertical GaN versus SiC.
MAURIZIO DI PAOLO EMILIO: Yes.
MAURIZIO DI PAOLO EMILIO：好的。
MARK DAVIDSON: So we’ll start with manufacturing. So, you know, at Odyssey, we have our own, I call it a microfab. So we have a 10,000-square-foot fab, and with minimal investment, we can get to 100 wafers per month, probably even more. So we can do quite a bit out of this facility with very modest investment. That’s one of the advantages we have as a company, because, yeah, we’re using all the same equipment as would be used by traditional silicon, but you know, the IP, the know-how, the expertise that we have is how to fabricate vertical GaN devices using this equipment. And that’s the hard part. There are plenty of papers talking about in theory how to build vertical GaN, but what I have is a team of really smart people who have figured out using this equipment, and they’ve figured out how to fabricate vertical GaN power devices and large devices.
You know, the product that we have in fabric now is our first sample. It’s a 65-mΩ you know, nominal RDS(on) device. So it’s by far not the largest. You know, we will be going to much lower RDS(on) in time. But it’s not just a tiny little test chip. It’s a chip that’s actually going to generate, or it’s going to conduct high power. And we’re doing that using this fully depreciated legacy semiconductor equipment, silicon equipment. What I’m really excited about is, we’re acquiring a new piece of equipment, actually two pieces of equipment. And this equipment was used by a Tier 1, non-analog, non-power, but you know, a Tier 1 semiconductor company in their own fab. We know the equipment was handled very well. We’re buying it for about 10 cents on the dollar. If I was going to buy the new equipment, this equipment new would be millions and millions of dollars.
We’re getting it for about 10 cents on the dollar. And that’s a huge advantage for us, obviously, on keeping our costs in control. But we’re able to use it, yeah, of course because of device size and all those things, but really because of the expertise of our team, we have materials and process people who have figured out how to use this equipment to build vertical GaN devices at scale, at eventually high volume, but right now being able to build large devices. So very excited about that. And yeah, like I said, that’s happening in our facility.
Twenty feet from me right now is the entrance to the fab. So that’s a unique advantage we have, being able to innovate that close to our R&D team, and of course, when we get to production, being able to control fab. As we’ve learned in the last few years, obviously it’s why the CHIPS and Science Act, being able to control source of fab is very important for the semiconductor business. So you know, it’s a huge advantage that we have.
MAURIZIO DI PAOLO EMILIO: So in GaN-on-GaN, so vertical GaN, I see in comparison with other substrate, GaN-on-SiC, so in particular that SiC is much more thermally conducted than GaN. Another thing is that the electromobility in GaN is much higher than SiC, but only in a lateral device. And so if you switch to a vertical device, you could lose some advantages in this case. What do you think?
MAURIZIO DI PAOLO EMILIO：所以在GaN-on-GaN和垂直GaN中，我看到與其他基板，即GaN-on-SiC相較，SiC特別在導熱性上高於GaN。另一件事是GaN的電子遷移率遠高於SiC，但僅在水平元件中。因此，如果切換到垂直式元件可能會失去一些優勢。對此你怎麼看？
MARK DAVIDSON: Yeah. I think on your thermal, you know, that’s a true statement. Of course, we’re able to thin our wafers to improve thermal conductivity, so that, you know, gives us, I’d say, it improves the thermal, to the point where, you know, I think we’ll be okay. And of course, time will tell. We’re getting parts packaged. But I think on the thermal side, we feel pretty good about that. Regarding your point on electromobility, you know, so we look at the whole equation for, we’ll call it figure of merit, figure of merit around specific on-resistance, etc. You know, and there are really three big factors: electromobility, dielectric constant, and then of course critical field.
And in the equation, mobility and dielectric constant, you know, they’re linear in the equation. And you know, yeah, there’s a little bit of what I would call, you know, puts and takes between GaN and SiC. The dominant factor is critical field. So relative to SiC, we have what I’d call a quantifiable advantage over SiC on critical field. But what’s really valuable for us is, in the equation, that factor’s cubed. So it’s to the third power, which is where we get significant advantage, you know, when I talk about that kind of 6× to 10× improvement, it’s really based on that.
So you know, like everything in technology, you know, it’s all about tradeoffs. And I think when we look at that figure of merit, when we look at the overall tradeoffs, you know, we have a pretty distinct advantage at the material level that we’ll take advantage of on our vertical GaN devices.
MAURIZIO DI PAOLO EMILIO: Which are, if there are differences, issues, in terms of integration, I mean, with respect to on-board drivers that we have on GaN-on-silicon, in this case, differences in terms of lateral and vertical ones.
MAURIZIO DI PAOLO EMILIO：如果存在差異，那就是整合性方面的問題；我的意思是，以GaN-on-silicon的車載驅動器來看，在這個案例中，水平和垂直元件的差異。
MARK DAVIDSON: Yeah. You know, I think, so in our structure, our vertical power FET, we will not be able to do integration. So we will be a discrete FET. It doesn’t mean we won’t have some sensing capabilities, but you won’t see us do what we see some of the lateral GaN companies do. So when you have a HEMT, you could build small, you could put your gate drivers on the same die, and we see that as an advantage they have, and really they need, because, you know, lateral GaN, let’s say GaN-on-silicon as an example, or GaN-on-substrate.
MARK DAVIDSON：在我們的結構中，我們的垂直式功率FET無法進行整合，我們會是一個獨立的FET。這並不意味著我們不會有一些感測能力，你不會看到我們做一些水平式 GaN公司所做的事情；當擁有HEMT 時，可以打造小型產品，可以把閘極驅動器放在同一個晶片上，我們認為這是他們擁有的優勢，而他們真正需要的是水平GaN，例如GaN-on-silicon。
You know, they tend to be fabless companies working oftentimes in the same fab as their competitors, and it’s pretty hard to differentiate at the device or process flow level. Right? The design rules are the design rules at the foundry. So the lateral GaN companies, you know, they’re looking to create differentiation through integration, as you say, maybe even with packaging, you know, those sorts of things, right? And there’s a lot of room for innovation there, and we see them doing some very impressive things.
And I think that’s why integration’s important on the lateral. But for us, you know, our roadmap is to build a vertical GaN FET, a discrete FET, maybe discrete FETs, right? Maybe integration could be in multi FETs, but we’re going to stay purely in a FET. We don’t see integration on our roadmap.
MAURIZIO DI PAOLO EMILIO: So Mark, in conclusion, you announced that you have achieved 1,200 V, and you are fabricating a first-generation engineering sample. So tell me, what areas do you think there will be opportunity with your technology, and what’s next in this case? So are you also providing development boards for testing, for practice, I mean, for your customers?
MAURIZIO DI PAOLO EMILIO：所以Mark，最後的結論，你們宣布已經達到可支援1,200V高壓，並且正在生產第一代工程樣品。那麼你認為你們的技術在哪些領域會有機會，在這種情況下，下一步是什麼？還有你們是否還會為客戶提供用於測試和實作的開發板？
MARK DAVIDSON: Yeah, absolutely. And thanks for taking note. So earlier in September, we made the announcement that we were able to build a 1,200-V vertical GaN FET, and that was really the stated objective of our company. We had already shown the world that we could build a 650- or 700-V vertical GaN FET, which is valuable in the market if we look at a SiC 650-V node, there’s a lot of businesses, a lot of growth there. But of course, in addition to that, we wanted to get to 1,200 V. So we’ve announced we’ve done that, and that was a significant accomplishment for the company.
MARK DAVIDSON：好的，當然。感謝你的注意。在9月初，我們宣布我們能夠打造1,200V垂直式GaN FET，這確實是我們公司的既定目標。我們已經向世界展示了我們可以打造650或 700V垂直式GaN FET，如果我們看看 SiC 650V節點，這在市場上很有價值，那裡有很多生意、有很多成長。但當然，除此之外，我們還想達到1,200 V。我們宣布我們已經做到了，這對公司來說是一項重大成就。
So yeah, so we’re fabricating samples. We’ll put them in packages by the end of the year, do our internal testing, and then pass them to the initial customers. I’m committing to three customers at the beginning, and the markets we’re going to start with are, there are some automotive companies. So we’ll pick one that we’ll give first samples to. I can’t go too wide, because we’re a small team and I don’t want to overwhelm the team with too many customer inputs. So automotive is one, electric vehicles, of course. We have a great relationship with an innovative electric motor company that’s looking, you know, they’ve moved from silicon to SiC and now they’re looking for us to take them to that next technology step with high-voltage GaN. So that’s another company that we’ll be giving initial samples to.
We see electric motors. I read a stat: About half the energy in the U.S. is consumed by electric motors. And honestly, I would fact-check that, but it’s a significant amount. And I was surprised by that, and I understand why the electric motors are so excited to come up with a higher power density, more efficient solution provided by our technology. Third market is renewable, DC-to-AC inverters for solar, photovoltaics, etc. We see that as an opportunity as well. And then, you know, we have some conversations going with some companies that built industrial power supplies. I have a call next week with a company that does power for high-rail and government applications, space, etc.
So, you know, we see a lot of electrification, sustainability, driving voltages higher, and obviously, we see that’s the problem we’re going to go solve. And then to your last question, for sure, I have a team we’re just building now of systems and applications, building evaluation boards, demo platforms, etc. So yeah, of course we’ll give discrete FETs to people to test, but we also want to give them some boards where they can just plug them in and see the performance benefits of our products. So evaluation boards, demo boards will absolutely be part of the go-to-market strategy.
MAURIZIO DI PAOLO EMILIO: Great. So looking forward to seeing vertical again in the near future. Thank you, Mark. Thanks a lot for joining us. It’s been a pleasure to have you in this podcast. Thank you.
MAURIZIO DI PAOLO EMILIO： 太好了，期待在不久的將來能看到垂直式GaN元件問世。謝謝Mark，非常感謝你加入我們。很高興在這一集的Podcast能邀請到你，謝謝。
MARK DAVIDSON: Maurizio, thank you for the great questions. It was a pleasure speaking with you. Thank you.
MAURIZIO DI PAOLO EMILIO: After talking to Mark, SiC is significantly better than silicon, but still limited, and still not as good as what vertical GaN is going to be. When we look at the capabilities of SiC versus vertical GaN, vertical GaN, according to Mark, gives about a 6× to 10× improvement over SiC. Mark said if you just look at the material properties, we see lower defect count and then less intersection because of vertical scale versus the horizontal scale. There are still challenges to solve, but as Mark said, we have our first samples about vertical GaN.
MAURIZIO DI PAOLO EMILIO：在跟Mark聊過之後，我們了解SiC明顯優於矽元件，但仍有其限制，並且達不到垂直式GaN的水準。當我們比較SiC與垂直式GaN的能力時，根據Mark的說法，垂直式GaN的性能比SiC高大約6到10倍。Mark說，如果你只看材料特性，我們會發現缺陷數量較少，然後由於垂直尺度與水平尺度相比，交叉點較少。雖仍有挑戰待克服，但正如 Mark 所說，將有第一款垂直式GaN的樣品。
They are using all the same equipment that will be used by traditional silicon, but the IP expertise how to fabricate vertical GaN devices using this equipment is a challenge. But according to Mark, they have to figure out how to fabricate vertical GaN power devices. They will not be able to do integration. They will be a discrete component. They were able to build a 1,200-V vertical GaN, according to Mark. There is a lot of business, a lot of growth there.