30% More Power, 20% Lower Costs: Scott Graybeal on Solar’s Step-Change Moment

Speaker A: What this technology promises is improved energy harvest, significantly improved energy harvest, 25 to 35 to 45% energy harvests, you know, depending upon what technology node we're talking about. And if you're trying to make money in solar, which people do, and that's why they're building it, dramatically improve cash flows.

Speaker B: Hardware to Save a Planet explores the technical innovations that are giving us hope in the fight against climate change. Each episode focuses on a specific climate challenge and explores an emerging physical technology solution with the person bringing it into reality. I'm your host, Dylan Garrett. Hardware to Save a Planet is brought to you by synapse. Make sure to click subscribe so you don't miss any future episodes. Thanks for listening. Hello and welcome to Hardware to Save a Planet. Today I'm excited to be joined by Scott Graybiel, the CEO of Calyx, to talk about solar energy. Calox's approach uses perovskites, uh, a material that can make solar energy more powerful, efficient, and cost effective. At a time when conventional photovoltaic panels are reaching their maximum efficiency, this could be a really big deal for solar. Calyx claims they can achieve 30% more power in the same space and 20% lower project costs. Calex was founded in 2014, and Scott joined as CEO in 2021. He has spent a lot of his career in clean energy and advanced manufacturing, with more than 15 years in the solar and energy industries, including seven years leading the energy solutions segment at Flex, a major global manufacturer. So he knows this space really well, and I'm excited to learn more from him. Um, Scott, it's great to have you on the show. Thanks a lot for joining.

Speaker A: Well, thank you for having me, Dylan.

Speaker B: So, actually, I was kind of scrolling through your background. It looks like you were working in Solar almost 20 years ago, um, and maybe before that. But that's when I sort of first noticed it on your resume, which is a probably sort of an eternity in the solar industry. Um, I'm curious what stands out to you as some of the biggest changes in the space since then?

Speaker A: Yeah, 2008. So I guess it's going on, yeah, 18 years this year. Um, I think the simple answer is the cost, clearly, because when we were starting out, crystalline silicon panels were expensive, you know, a couple of bucks a watt. But then when I talk to the real veterans in the industry who've been in like 30 to 40 years, you know, they remember $10 a watt. And so that's been a big standout. And I don't think anybody back in 2008, when I started, really had a concept of what the scale could be. I remember hearing I started off in the thin film silicon world, which came about because of really expensive polysilicon prices. When it was like $350 a kilogram. And thin film silicon was going to be the answer to that. Well, it was, it was funny because, you know, we would talk about big factories back then, a big factory. I think Sharp had a, uh, factory in Japan that was 160megawatts and we were just ooing and awe over how big that was. And you know, now we have a pilot line that's 50 megawatts and we're scaling that to 100 megawatts by the end of this year. And that would be considered a huge plant back in 2008. Today it's, you know, it's a demonstration line. So it's, it's a very different world when you think about the scale. And I think just the, the fact that solar has become so mainstream in most geographies is really been remarkable to see. And it's been faster than I think anybody had anticipated. Um, so the pace at which this industry evolved was, was quite dramatic. But I think the challenge really is, is that it's all been along the same kind of monolithic node. You know, panels look pretty much the same as they did in the 1970. Did have some pretty funky looking panels with round wafers and whatnot around cells. But, you know, more it really hasn't evolved much. And when you look at how other technologies have evolved, where we've seen increases of processing power, et cetera, you know, there really isn't an analog that we've seen in solar until recently. And I think that's what's going to make the next 10 to 20 years exciting is to see these changes come about from just how we think about deployments of solar and the materials that we use in solar. But yeah, it's a, uh, it's been interesting to kind of see this industry evolve and I had the good fortune of being everything from the manufacturing side to the equipment side to doing a little bit of project development. Um, so I got a pretty good view of how a lot of these different pieces fit together over that time. That uh, it's been, it's been a journey.

Speaker B: Yeah. And, and like you said, it's been, it's been kind of a broad view of, of solar itself. And then even outside of solar, it looks like from at least your time at flex kind of energy more broadly. I'm Curious. What was it about, um, were you specifically looking to get more deeply into solar again, uh, after Flex? And then what was it specifically about Calyx that attracted you as the solution to focus on?

Speaker A: I took a short break from leading the energy business at Flex and we took that from about 300 billion to about $2 billion. And we did that organically and energetically, like the next Tracker acquisition for example. And then um, you know that was a, ah, it was a fantastic ride in many sense, in many ways. But in other ways it was a real challenge because we saw some of the larger bankruptcies, you know, had occurred in around that time, like Sun, Edison, et cetera. But amongst, you know, what were challenging times are also this maturation of the industry. And so I took a short break into construction technology for a bit because for me that was also very much a kind of a mission driven approach. How do we make housing cheaper and more plentiful? Um, and so that was a little ah, segue. But I stayed engaged in the industry as an investor. So I had as an angel investor invested in startups that were in the solar space and it was always calling on me again. And I got the call from the venture fund that was backing the company at the time and is still with us. Um, they asked me to take on this new opportunity. And what was compelling was I had not seen a step function change in performance like this type of technology Perovskites brings to the table. You know, we used to grind it out trying to get 1, 2, 3, 4% improvement in energy harvest on a relative basis. And to have something that could promise a 20 to 30 to 40% improvement on a relative basis was just almost unbelievable. And um, it took me a little while to get my head around it and I brought in my own experts and I folks that I respected that were cell scientists and I said, is this real? You know, what do you think about this? Because I couldn't believe the math. And the takeaway after they, the tires was yeah, it looks real, um, it's going to take a lot of work, but it's there. And we were at that time the company was on 2 inch by 2 inch substrates. And now five years later we're producing and shipping 2.1 square meter substrates. So there's been a huge leap forward in terms of the size at which we're producing these things. Um, but what was attractive was just the possibility of really changing that monolithic, let's say architecture I talked about previously and really bringing something new to the industry and allowing it to expand further. You know, there are projects today that don't pencil out effectively using the current state of the art technology. And we are running into a ceiling. Um, I think we will be pleased to see pure play crystalline silicon technologies reach maybe 25%, practically speaking. And I think that Perovskites will enable us to really legitimately talk about 30 and 40% efficiency products in the future. And that's the promise of tandem devices, when you're able to slice the spectrum up into bits so that you're able to more easily digest the energy that the sun is putting out. And that's exciting. You know, that's really a shot in the arm. I think for anybody who views PV as a technology that may be on the fringes. This type of a conversion efficiency that we're talking about drops costs so dramatically at the same time, really fills a much needed requirement that we see today. Because all we're talking about now is about energy costs rising due to the pressure of data centers, et cetera. And our view is that you couple high energy density solar with low cost energy storage and you have a dispatchable solar capacity that can support the needs of the future. So without this technology, without Perovskites, we really don't get there.

Speaker B: Yeah, just as a little tangent, what, how were we kind of eking out more and more efficiency using fundamentally the same technology with uh, kind of silicon based tv?

Speaker A: A lot of, a lot of it was geometry, believe it or not. You know, making the dead space that was on a module less and less. And there were other approaches where some people were shingling, for example, where they were cutting wafers and gluing them to each other so that you could eliminate the dead space between the individual solar cells. Some companies have really figured out how to lay up solar cells very effectively to remove that. And then you've had some incremental innovations, I'll call them, where, like something like a heterojunction device where you're putting a thin layer of amorphous silicon on a crystalline silicon cell and doping it appropriately, you know, that would give you maybe 1 or 2% improvement in efficiency. You know, you see like the Topcon technology which, you know, uses different oxides and, and takes advantage of other, let's say interesting, uh, physics properties of silicon to get incrementally just a little bit more out. But that's a lot of capex. It's a lot of effort and it's getting more and more expensive to get those increased efficiencies out of silicon. You Kind of hit the point of diminishing returns. Whereas, you know, these technologies, perovskite technologies and ours in particular, um, you're able to get pretty dramatic leaps forward and there's a lot of headroom left. You know, we're just really getting started in the commercialization of this technology. And so we've got still a lot of dry powder and there's a lot that can be done to continue to optimize perovskites.

Speaker B: And. And, uh, let's see if. Yeah, I'm just trying to remember. Calyx was founded in 2014, I think. Um, what was their role in Provskites? Uh, you know, are they kind of pioneering the field in some way? Do they have a particular angle that others don't? Um, sort of. What. What's going on in that space?

Speaker A: Yeah, so the company really started with the idea of how do you boost the performance of crystalline silicon technology using different material systems? And they tried a lot of different things. Um, they tried sigs, they tried, uh, other thin film approaches, silicon nanowires. I mean, there's a lot of different ways they did. That was really John's vision is that, look, unless we go to some type of tandem structure, it's going to be harder and harder to get these single junction devices to really perform where they need to be. Um, and then in about 2017 or so, um, the company aligned with the board on a, on a shift in direction to go and look at these perovskites. They started really, you know, the discovery was like, in 2009 in Japan, and there was a lot of work, but back then, you know, these devices lasted seconds. You know, it was literally, you produce a perovskite and measure it super quickly because otherwise it was going to degrade. That's no longer the case. Now we have lifetimes of years. Ah, and that's a big step forward. So when the lifetimes really started improving and the cost window looked like it was achievable, um, then that the company made that shift. Um, the innovations that we've brought to the table are around the chemistry. So everybody's got a slightly different approach, uh, from a chemical standpoint. Our process, though, is quite a bit different in many ways. So you have to look at these devices as not just being perovskites, but they're a stack. And so there's certain elements of that stack that we have proprietary technologies that ensure that we have the lowest cost pathway to produce perovskites, and, um, even lower cost than players in China. And that's because we eliminate certain capex that is otherwise needed by other players. We eliminate that completely and use low cost methodologies and process conditions to make these devices very quickly. And that's been a key differentiator. Our focus has been on the scale production of these devices. And John and our founder very early on limited the chemistry window that the team was allowed to explore to really um, I would say chemistries that were going to be scalable where we didn't need some exotic dibublio muctate that was going to be super hard to produce. You know, we could use largely commodity chemistries and use them effectively in scalable processes so that we could make these devices commercially quite successfully. And you just as an example, excuse me, once we scale this site, you know, it'll be quite profitable. Um, once we get to our 100 megawatts we'll be completely in line production. Um, and you know, considering where the, we know we can sell the product at and where uh, what it's going to cost us, you know, it's exciting to actually run a solar business in the United States that can run a profit at a very low capacity. And likewise that's helping us our international place as well. You know, we're, you have letters of intent for very large deployments of other factories and other geographies which we'll probably talk more about as the year progresses. But those are enabled by all the hard work that's been done here in California to make a scalable, low cost, high efficiency, long lasting perovskite system.

Speaker B: Hmm. Um, can you explain that? You've mentioned that it's a tandem system a couple times. What does that mean?

Speaker A: So let me first start off by talking about why people do tandem devices in general. It's the way that you can optimize the sun that you're getting. And so what you do with the TAM device is you split the spectrum up into chunks. And uh, near the band gap of a given material, the material itself is much better at absorbing photons from the sun and converting those into electrons that we can use. So we call that the band gap. It's the energy at which they're promoted from the valence level to the conduction band. They're semiconductor photoactive materials.

Speaker B: Mhm.

Speaker A: And so if you've got. That's right, that's right. So in specific terms, silicon has a band gap of about 1100 nanometers, 1.12 electron volts. So anything that it absorbs that's more energetic and that 1100 nanometers by the way, that's in the infrared part of the spectrum. Anything that's more energetic, like visible light, it still absorb. It absorbs it, but it's indirect band gap semiconductor. And those charge carriers get promoted to levels that are less efficient, and then they kind of decay down into. And I'm probably explaining this incorrectly, this is how I understand it. But they, as they decay down into the conduction band, they give off heat. So you'll see the cell temperature for silicon is warmer than what it would be in a tandem device. So when we talk about tandem devices, they're actually cooler because the perovskite absorbs around 750, and that's where it's optimal, which is in the red part of the spectrum. So red and visible light is more efficiently absorbed by the perovskite. And you pass down all of that red and infrared light to the silicon, where it's pretty efficient at absorbing it. So this is how tandem devices generally operate. So you're finding you pick these different band gaps, these different parts of the spectrum, and say, hey, I'm going to be really good at absorbing blue light, I'm going to be really good at absorbing red light. I'm going to be really good at absorbing infrared light. And by cutting it up in those little sections, and you're able to convert sunlight to electricity much more efficiently.

Speaker B: Got it.

Speaker A: So that's what we mean by tandem.

Speaker B: And you described it as a stack. So one layer is kind of capturing a certain portion of the spectrum, uh, that it absorbs most efficiently, and then the next layer, uh, absorbs light in a different. Is that the right way to think about it? Yeah.

Speaker A: The stack I was referring to was the stack just for the perovskite system alone. And the way our architecture works is that we have this perovskite layer, which is deposited on the COVID glass that goes on a module, so it's on the underside of the COVID glass. Then we have a lamination material which allows. Gets everything to stick together, but also allows the light to go through and at the same time electrically isolate the crystalline silicon from that cover glass. And so now we have two electricity generators in one footprint. So we have the perovskite cell, got, uh, an insulation material in between, and then the crystalline silicon cell, which is converting that, that, uh, deep red infrared light into electricity. And so they're connected electrically in the junction box in parallel. So the voltages match, but you can add the current together. And so the stack I was referring to was the stack for the perovskite system, where you have a uh, transparent conductive oxide, something called the hole transport layer. Then you put down the perovskite, which is kind of a complicated system of different chemistries that come together. And then you've got an electron transport layer and then you have another oxide and that's a conductive oxide. And that's what we ship to customers who make the modules out of what we call active glass.

Speaker B: Okay, help me understand the significance of this. So, uh, a step change in efficiency in kind of energy we can produce with a certain amount of uh, sort of square area, uh, materials, potentially lower, uh, cost. What does this mean for, I don't know, our clean energy transition for the solar industry? Does it open up new use cases? Um, what's the significance?

Speaker A: I like to go to the economics, but there are use cases you touched on are pretty fun too. So let's hit both of those. Um, so the economics are what really captures, I think the minds of folks that are interestial solar today that are building these projects. And what this technology promises is improved energy harvest. Significantly improved energy harvest, 25 to 35 to 45% energy harvest, you know, depending upon what technology node we're talking about. Um, and if you're trying to make money in solar, which people do, that's why they're building it, um, dramatically improved cash flows. So between like 45 to 55% improvement in project cash flows over a 25 year solar power project. And then a lower levelized cost of energy is the flip side of that. So let's say you plow back. Instead of that profitability, you just lower your levelized cost of energy. Well, that lower levelized cost of energy now opens doors for projects that would otherwise not pencil out. So you could be in conditions, let's say for example. Well, it's a fairly low cost region. Great. Well, we couldn't get in there with crystalline silicon technology, but if we do a, uh, tandem technology, we can actually build the project successfully. Um, I think in applications like we were talking about before, like data centers, it's huge because now you have the ability to either instead of building a massive solar farm, you can be maybe 30 to 40% smaller and still get that same energy. Or you could build the same size and just get more energy out. Really up to you. Combine that with low cost energy storage, you effectively have kind of a dispatchable system. And the ratio of that really goes like 1 to 5 to 17 is kind of how the math works. So for every, let's say megawatt of demand that you have from a data center, you'd need 5 megawatts of solar, then you need about 17 megawatt hours of energy storage. And that's generally how that that ratio comes together. Um, and the cost of energy storage is plummeting and with innovations like this, the cost of solar can be sustainably lower. You know, we saw a pretty dramatic drop in module prices in China in particular hyper competition markets getting shut off to them, whether tariff or non tariff measures internationally, which cause those prices to plummet. With Topcon technology, for example, could be purchased for 8 to 9 cents a watt. Well, that's changing. Prices are going up and we think they're probably going to set a lot around 15 cents a watt as the industry really recovers and mothballs old capacity resizes itself and markets kind of figure out what they're doing from a, uh, regulatory standpoint. So that was an aberration and we knew it was going to be an aberration. But longer term, you know, we'll see prices in that 15 cent range for at least the next couple of years. Um, well, that being said, you know, that then how if you were to go and take those same said modules and include a perovskite top glass on it and our active glass as an example. Well, uh, maybe the module price per watt goes up by a couple of pennies. So now it's 17 cents a watt, but now you have 30% more energy harvest, so it quickly pays for itself from that standpoint. And now you talked about use cases. The use cases for these products are amazing because you also have the opportunity to deposit them on flexible substrates. And that's something that we've kind of been working through. And the flexible substrates really open up a lot of different markets and maybe rethink how we do terrestrial pv. You know, maybe there's a future where we can just print out rolls of this solar panels if you will. And they're so cheap that you really, you know, a lot of the other considerations go away. You really don't care if it lasts 25 years. I'll just go down the Home Depot, hit the button and print out some more. I mean it's possible, right? And that's, and we like, we love to think about those things, you know, integrating photovoltaics more effectively into automobiles. Um, what an exciting thing to be able to do, to be able to provide an automobile manufacturer a film that they could put maybe laminated inside of the glass that goes on a moonroof and say hey, there's. There you go. It's a semi transparent array that will generate electricity, get you a few extra miles on your electric vehicle. Uh, space applications are just through the roof. Perovskites do very well in space. Um, we've worked closely with Caltech over the years to prove that, you know, they have, uh, the ability to do a proton in bombardment. Um, and we've been able to prove it. Plus we've also put material in space. We've been in low Earth orbit and our Perovskites, even after a solar storm, performed extremely well. These, these materials heal themselves and that healing factor is something that silicon doesn't exhibit. We even see this in the terrestrial space where you'll see a diurnal response where, you know, you'll have the, uh, perovskite during the day will be producing and maybe trails off towards the end of the day and then recovers completely at night. So you see this cyclicality. So that's also something that has not been considered before in photovoltaics, where a lot of the tests today are geared towards silicon. They're all geared towards silicon, frankly. And so these types of phenomena are not accounted for, but they exhibit themselves quite well and quite obviously with perovskite. So there's a lot of cool things that will come out. We may even be talking about solar embedded into clothing. Who knows

Speaker B: on the economics point? Um, uh, I was curious. You're talking about there are projects that pencil with Perovskites that wouldn't otherwise. Pencil. Is it right to be thinking about that as if in the sense that, um, those regions would otherwise be served by natural gas or something, or coal or something. Uh, that would. Pencil. But in this case we can replace that with solar or what. What is the alternative for places like that?

Speaker A: Yeah, that's a good question. I think it's going to depend case to case, and it's going to depend where you are geographically or close. The coal. Do you have availability of natural gas in many parts of the world, you just go without, right? So if you don't have, if you don't have natural gas or you don't have coal. Okay, well, we're just going to have a restricted grid, I guess. You know, the US case, for example. You know, when I, when I think about those use cases, yeah, there's always the optionality. But one of the things that we've done a lot of work on will be hopefully publishing a white pa, you know, late this quarter, early next quarter, along with some energy storage folks, is really this idea of what's it look like to have dispatchable solar when you're talking about having high energy density generation from perovskites and low cost energy storage. Um, and you can see that for example in Europe that we, we accelerate the obsolescence of natural gas dramatically. So we could be, in three years we could be looking at the obsolescence of natural gas in Europe, even in gcc, uh, we could see that in four years and in the US it's going to take longer because we highly subsidize natural gas extraction and transportation. So either through business structures like master limited partnerships or through direct incentives such as percent depletion deductions and intangible drilling costs, natural gas is highly subsidized here. But with that subsidization level, then you're looking at the horizon for natural gas obsolescence is really in the late2030s. So if you're planning out a natural gas project and if you know it's going to be obsolete in 10 years, that may cause you to pause and think about it. Um, whereas we have kind of a modular building block almost like Legos. When you think about what solar and storage can do together, um, it's a very different paradigm. So when you see that projects aren't penciling out, it's usually because it may not meet the horizon for the returns for the developer or the independent power producer. And so great live use case, we're seeing today the limited availability of Topcon technology, which is a couple of percent more efficient on absolute basis than the predecessor Perc. So developers were signing projects with Topcon in mind. Well, that became a very messy legal battleground as companies were beginning to sue each other over intellectual property rights associated with Topcon. So what replaces that? Well, what we've been able to prove is that you can integrate our technology with Mono Perk and it'll perform better than Topcon. So we're shipping today as what we call our Calyx prime product. It's a beta product, so limited release, 25 customers worldwide, we're shipping the product to. And what that promises to do is to take what would be a 21% module and makes it a 24 to 24.5% module, which by the way is a full percent on an absolute basis. A full percent more efficient than Topcon at 23 and a half. So you could take this legacy like 10 year old solar technology and make it perform better than the current state of the art crystalline silicon technology at the same price here in the Us, that's what's exciting. So what we were paying before. You're going to pay or it'll cost whatever the module company's cost was before, it'll be that whatever they choose to charge the developer, that's up to them. But in terms of our brand promise, it's like, look, we're going to take a third of the value and you, Mr. Module Company, Mr. Developer, you fight it out for the other 2/3. So whatever you think that fair split is, and we think that that's fine, so everybody could take a third and we all win. Mhm.

Speaker B: So what's the catch?

Speaker A: Yeah, what's the catch? Well, I think that um, we need to scale this and we need to scale it quickly. I think as an industry, um, we're excited about it. We've, as I mentioned before, we have offshore litters of intent, well over 10 gigawatts of manufacturing capacity, a portion of which that's going to start this year and then with the line of sight for production next year, uh, late next year. And I think we need to get there, we have to have the will to do it. Um, and I'm finding that there's a lot of interest and other GEOs for this technology outside of China. Um, you know, I think the catch really just comes down to, you know, like any new technology we're going to have to really deploy it, understand it, take those lessons and plow it back into the next product revision and go along that learning curve.

Speaker B: Right, yeah. Which solar has done so famously.

Speaker A: Yeah, yeah.

Speaker B: Um, and, and when you talk about producing it at scale, are you producing a material that's then supplied to solar manufacturers? It actually sounds like you're producing more of a, a, uh, product that goes into panels. Can you say more about that?

Speaker A: So what we do. So every solar panel has, or every solar module has a top glass. So replace that top glass with active glass. So we take glass in our factory and that's what we deposit our system on. And so we have the capability of depositing on large sheets of glass about 2.1 square meters this um, material stack. And by the way, there's no length limitation. Um, we know that that's smaller than utility scale today. Utility scale, they're typically 2.3 meters, 2.4 meters and we're about, you know, 1.95. Um, I know, but this is the size of the tool that we had. So we started with used equipment that was at a certain size. But the next factory we have will be able to do you know, up to 3 meters in length easily. So length isn't the issue. You got to get the width right. And let me tell you, these, uh, everybody thinks this is easy. You know, just go get a turnkey line. No, this is hard. It was three years of really hard work to get from the moment where it was like, hey, I think it's. I think we see a band gap here. I think it's photoactive to. Okay, now we have a product that we feel comfortable shipping to Beta customers that's at a, you know, 15% minimum efficiency, which then turns into that 24. 24 and a half percent efficient tandem module. Um, you know, our record today as we published and we put out LinkedIn, because normally we don't get too explicit, but 16 and a half percent right now, and our full release product will be 17. 17 and a half percent, really? 17 will be the threshold. And that'll come out in June. And we think it's going to be faster because I've been blown away by how quickly this man. Once we kind of figured out some things and it took a while, you got to do a lot of experimentation, but that efficiency shot up dramatically. And, um, it's not just one thing. Somebody asked me recently, I think it was an investor. So what was the breakthrough? There was no breakthrough. It's a grind. The breakthrough is that you just got to keep running and you can't let your site go idle. You can't make one good panel a week. That's impossible. Um, and it's naive to think you can do that. You have to run dozens and dozens and dozens of these things a day. And to really get the learning. And today, we will be making by the end of this year, about 800 of these things a day. Seven to 800 a day. So that's where the learning comes in. And it's all Wright's Law. Um, Theodore Wright, back in the 1930s predicted these and mart innovation was being able to take that mathematical model and apply it to parametric performance. And so, with a high degree of confidence, we could tell you if we produce X number of devices, this is what the efficiency would be. And people doubted us on that. But we proved it. It's true. We run X number of devices, we will get to 17% efficiency. We showed it with our model so far, and we're proven the same thing to be true with our reliability work that we're doing, the durability work we're doing on this technology. So, uh, it's all just. We're kind of taking the lessons that China taught us again through the crystalline silicon, then applying them back into what we're doing and what we really call, look at this PV 3.0, you know, this new where I think the whole game gets reset. It's really not a China conversation anymore. It's really about, you know, where can this be done and who are going to be the technology leaders and in each geo.

Speaker B: And, and uh, as you're kind of achieving these improvements, what is it exactly that you're iterating on? Is it, is it sort of process optimization? Is it material science? Is it, you know, supply chain? What, what's actually changing?

Speaker A: Yeah, there's, there's parallel efforts obviously. So R D really focuses on um, you know what, it's going to be the next gen product. What we have today suffices for our next four generations of our product that we have on the line. And it's coming down to simple things. Well, simple things are often complicated. Everything from defect reduction to tighter process control, optimized process conditions. And that's what we're seeing the gains. And so we can easily get I think to a 20% device, top class on the current node that we have, the current chemistry that we have, um, and there will be advancements that R and D works on to address even higher levels of efficiency. So that really puts us in a place that it's really coming down to just process optimization. And uh, the supply chain aspect of things has been absolutely critical. Um, when we got our first round of financing, I told the team, I said guys, cost reduction starts today and we have to be mindful about where we take the technology and who our partners are going to be. And that's been incredibly important. Um, so we professionalize our supply chain team quickly where we have international supply base. Uh, we work with companies all over the world to support what we need and also try to develop new suppliers. And that's been enabling us to get our direct materials costs in line to what we predicted again using our favorite guy, Theodore Wright. Um, it's been, that's how we've been able to do it. But that's just another problem. You got to grind it. And uh, one thing I learned out of Flex was how to grind profit out. And so you know, it's really a lot of blocking and tackling. I hate to say that there is no, there's no silver bullet, you know, it's hard work.

Speaker B: Yeah. So then your business model, you're selling, um, you're selling this covered glass or sorry, this Active Glass, I think you called it. Um, and I saw, I read somewhere your, your strategy is to partner with incumbents. Um, I'm just curious if you could say something about that as opposed. I suppose the alternative would be to be to compete with the incumbents, kind of making your own modules and this kind of thing. Um.

Speaker A: What, yeah.

Speaker B: How did you pick that approach?

Speaker A: Well, there were, there were lessons. So one of the things I realized when we did the deal with Next Tracker way back in 2015, the biggest gift that Flex gave to Next Tracker was its balance sheet. Because they went from a single customer, um, startup company, 45 million in revenue, and a year later, 450 million in revenue. And that was because you had a large balance sheet player that could sign parent guarantees that unlock the possibility of working on all these different solar power projects that otherwise were closed off. And so thinking through the same thing, you know, one of the, one of the things I was trying to create a synthetic version of that with Calyx. And so the concept was let's work with large balance sheet, well known solar module players that can incorporate our technology into their platform and then the developers still working with the same module company and get the warranties coming from that same company they worked with the last decade. And so. And I also figured, you know, these module players, they're not going to give us any slack. You know, they're really going to push us through our paces, you know, more so than I think any third party ever would. You know, third party lab. I mean these guys will, you know, really, um, like I said, put us through our basis. Exactly. And so getting qualified with these guys is a big deal. And, but they also see an economic benefit from it too, that offsets the risk considerations. And so that enables us to get into projects and into, let's say, uh, you know, large residential installers that otherwise would be shut off. To us. My biggest concern was being a module company because I think people underappreciate how hard it is to be a module company. I mean I made, you know, we had a couple gigawatts of module capacity when I was at Flex and you have to be at 99.8% underlying yield to be world class. And that takes an incredible amount of discipline and it is a messy process, but you have to have that level of control. And I didn't want to solve two problems at the same time. You know, getting perovskites on glass. Right, is hard. Uh, compound that with trying to scale up a module factory. And I've had module Factories that scaled beautifully. I've had module factories that took two years and it's like, it's a lot. Even though you could say, oh, it's turnkey. It's never turnkey. There's no such thing as turnkey. Um, I sold turnkey lines for three years. It's not turnkey. Nothing's turnkey. I mean, it's really hard and it's more capex, you know, it's more capex. You know, whereas we can go, we could partner with a module company and if they're the right module player, then maybe we have a conversation about a potential joint venture or, you know, the same thing applies to glass companies. Let's do a JV with a glass company and then, you know, we can have Perovsky coated glass. The Calyx Active glass goes out and supports all these module companies and that's a good place to start. Now I say start because eventually perovskites will get really, really good at perovskites. And then maybe perovskites even replace silicon in the future. Maybe we start talking about dual junction and triple junction perovskite only devices on flexible substrates. I mean it can get really crazy really quickly. So maybe in the next 10 years solar's going to look a lot different and um, that's exciting. A lot different, a lot cheaper, a lot faster. Yeah, gotta start colder though. And that's where this partnership concept really comes about. Maybe I'm wrong, but, you know,

Speaker B: um, you've mentioned China a couple times. What, what's preventing China from jumping into this game and, and, and kind of doing the China thing with this, you

Speaker A: know, um, China is in the game and uh, there we have esteemed competitors in China, which, you know, I will never disparage. I think the good work that's been going on over there. But I will say this, is that there's a tendency for these companies to be module players and they want to compete against the incumbents. Um, they don't quite have the same partnership strategy that we do. And plus we're pretty, uh, we're relentless about our IP filing strategy. We file everything internationally and uh, we will defend our IP if we feel it's being infringed. Um, and so, you know, uh, if somebody can do it faster, better cheap than us, then great, they deserve to win. But we don't think they can. But our strategy is to be everywhere outside of China and uh, so far so good. You know, even though we're just really getting started on this journey, you know, that, that that activity is continuing. Um, but what's interesting about this case, when you compare it to what was made the crystalline silicon industry ship so quickly. We don't have massive capex requirements like you did in crystalline silicon, where you had to go build a giant polysilicon plant and be close to a coal mine to provide you the energy that you're needed to produce polysilicon. So that's gone in our world. You know, nowadays it's really, I've got glass, which make anywhere in the world, really just have some good sand. And now I've got the ability to just take off the shelf chemicals and synthesize them in a certain way that makes these cells. And so the capex to do that versus crystalline silicon is like a quarter of a crystalline silicon completely integrated line. You gotta think about crystalline silicon as being everything from quartzite all the way through the module. And that's the, that's really the entire value chain. Now we have a really simple value chain. That's okay, I need a glass guy and now I need some chemical guys and then I'm good. And then I can go and produce these things. And so that resets the value of having a highly concentrated manufacturing presence. So you don't need to have these massive polysilicate producers somewhere in uh, Outer Mongolia that are just producing all this poly. It's now a distributed manufacturing model. So now you've taken away one of the advantages. One could have been a labor or capital advantage. Now you decrease the cost of the capital significantly. So now that's a less of an advantage. Um, case in point, we don't even have a clean room. That's how different this is. We have clean room sections that are dedicated for certain processes. But we don't believe that there's a need to put together a complete clean room. Um, and we've proven it does. It's unnecessary. But it's anathema I think how some folks are thinking about this stuff, but we. It's not needed. Um, but that also shows you, hey. That well now you can open this up to a lot of different manufacturing locations. We're in an old pet food distribution warehouse. That was the business before we were here. So now you're looking at all this available warehouse space. Looking at it very differently. Maybe this is high tech manufacturing space. That's, that opens up a lot of possibilities. And the labor arbitrage is not available anymore because these lines are highly automated, highly automated. You know, we can run a uh, 2 gigawatt line that would take roughly maybe 5 people per shift, 20 people can run that line, not including, you know, HR and whoever you need for like the, you know, your rock functions to support the operation and management whatnot. But that's pretty light. So there real now there really isn't much in the way of any sort of labor arbitrage opportunity. So now we would drop capex where they had a significant advantage. We've addressed the energy consumption, we have 1/20 the energy consumption of crystalline silicon production we don't have. And so that CAPEX also comes back to the fact, okay, now you have more, now you're dealing with lower capex, so now you have more sources of potential capital. We don't have labor requirements that are going to be untoward. So now that kind of takes all the advantages that China had in crystalline silicon and reframes it for this new technology.

Speaker B: You uh, you were talking about the future a little bit ago and where this might go. I'm curious where as you look, as you sort of look in your crystal ball down the road, where do you see, how do you see solar sitting in the kind of energy mix of our future? You know, we've got a bunch of energy or a bunch of effort and money going to like geothermal and fusion and you know, wind, although maybe not so much these days. You know what, like what, where, how do you think solar is going to, going to evolve in the future as

Speaker A: part of our mix? I think it's going to become so ubiquitous we're going to take it for granted. And what drives ubiquity is cost and performance. So those two things and that's what this technology enables us to do. And I say that, you know, with all seriousness because we, I, you can see where the manufacturing footprint for a gigawatt of this stuff may be no larger than you know, 200ft by few hundred feet. So something that we would normally have seen is like maybe a commercial printer, you know, kind of uh, footprint. And I think it'll be so inexpensive that we'll start talking about cents per kilowatt as opposed to cents per watt. And that would be amazing, you know, and it, and that's going to really require thinking about solar a bit differently. Maybe it's not a 25 year product. Could we do it enough that you really don't care? It's like okay, yeah, you probably could, you know, could we just roll something out and stick it to a surface and now we just figure out how to go connect it and you know, that's not our wheelhouse. But we'll work with partners on that. In fact, we have a next generation module, um, program with another partner that's just kicking off. And so I think the sky's the limit. We just have to be willing to go and look at this technology and say it can look different. We can support the industry as it sits today, no problem. But let's challenge ourselves to think about maybe how we connect things electrically is stupid. Maybe the way that we're installing terrestrial today just needs to be completely turned inside out. Maybe we think about what's it look like. We talk a lot about space nowadays and I hope we talk more about it in the future. But being able to print out in situ what they're talking about, data centers in space, I don't think that's a crazy idea. Especially if you can go print out in situ solar panels to replace those that are damaged by micrometeorites and just have your robots out there stretch them out and say, okay, we're ready to go, we're back up running five nines. Those are all legitimate things for us to be thinking about. And my dream here is that it does become ubiquitous. Now the question becomes, well, how are you going to make money at that? Look, we'll make money at it, we're going to figure that piece out. But let's have that as really kind of the end state where it's simple access globally. Some people probably wouldn't like that. Particularly if you're in the O business, you'd like centralized production. I get it, that's how you're making money. Sure. But you know, the sun's been out there. Let's just start using it better and more freely. And there's, there's ways that we can start harnessing technologies like this in order to really achieve that.

Speaker B: Love it. Scott, I have three questions for you that I end every episode with. The first is, uh, how optimistic or pessimistic are you about the future of the planet and why?

Speaker A: I'm, uh, I'm, I'm pessimistic in the sense that I don't see political leadership aligning around science. And that's what concerns me greatly. I hope this is a generational issue and I'm hopeful that the next generation, like, you know, I've got kids that are 26 and 30, um, they think differently and I love that and I'm hoping that they can, they start to re, engage with science on a much more meaningful basis and use science as the basis for decision making as opposed to, you know, my Buddy at the golf course said ah, this was a good idea and ah, I, I would love to see that. And I think we, we've got to get back to rational science based conversations. I think if we can do that culturally then we're in good shape. Uh, if we can't, well, okay, it's going to look, I think the planet's going to be just fine. It's going to be civilization that's in trouble. It's going to plan will shake us off like fleas and it'll be fine.

Speaker B: Yeah. Who is another company or individual doing something to address climate change that's inspiring you?

Speaker A: Oh, you know, I got to say, uh, you know I've known Dan Sugar for a long time. Dan and Alex and that crew at Next Power, um, you know, they just pinned the throttle and uh, they are die hard believers and again with science. Right. But I love what those guys are doing. I think that they, they've got the, the right focus, they're going out there the uh, problems that we otherwise would ignore, you know, and they're applying really innovative solutions to those problems. And I love to see that. Um, and there's a couple more, you know, I think in the energy storage space there's a lot of promise. I just want to see more of that. So. Pretty excited about that as well.

Speaker B: Yeah, like new, new, new uh, battery chemistries and things or what, what catches your eye there?

Speaker A: I think new battery chemistries are always compelling. I'm always really fond of, well, let's go and get as much as we can out of this technology node. So for example, you know, we were one of the miles of flex, we were one of the first to really embrace LFP alongside enphase. Then phase kind of came out with their battery with lfp and I'm scratching my head like, wow, everybody else is doing nmc. Well then there was this LFP that was coming out. Then we started pioneering that for stationary storage. In fact, there's a flex energy storage, um, deployment at the National Wind Development center in Boulder that all the politicians used to love to get their picture taken in front of. And that was a Flextronics made energy storage system. Um, and so I think that there's, there's figuring out how to do it cheaply and at scale is largely a choice here in the US I think that that's one of the things that we should probably just decide we're going to go do as a country and invest in LFP energy storage. And then the next generation on top of that um, and there are some interesting chemistries out there. You know, anybody can run around with a, you know, a plus two charge as opposed to a plus one charge naturally has a leg up. So here's to all the zinc guys out there. So I think that's kind of interesting. I want to see what happens with sodium batteries. Uh, that would be super cheap. Right. There's some interesting things that are going to happen, and it's just coming down to scale.

Speaker B: What advice do you have for someone not working in climate today who wants to do something to help?

Speaker A: Yeah, I think people need to stay engaged in climate science in one way or another. You know, be naturally curious about that. And from there, you kind of let your heart dictate where you want to go. And it could come down to the ballot box. Um, I would say really just be another voice out there that's acknowledging the truth and saying, look, you know, these are challenges, these are problems. You know what, this is something we really, really need to get ahead of. And, um, I, you know, I'm not. Oh, this sounds terrible, but a lot of discussion around micro behaviors. You know, oh, should I get plastic or paper at the grocery store? You know, those are not needle mover events. But what are needle mover events is you get the people in power that are able to make the decisions that are going to be beneficial to the planet. That's ultimately it. So let's, let's harness our democracy appropriately and acknowledge the fact that this is a challenge, that we have to go after it. So I think stay informed, be fearless, uh, learn as much as you can. And I think that's always just the basis for just about anything. Be a responsible citizen.

Speaker B: Love it. Scott, that was awesome. Thank you. I learned a lot. Thanks for everything. You're doing well.

Speaker A: Thank you, Dylan. Really appreciate you having me.

Speaker B: Hardware to Save a Planet is brought to you by synapse. To find out more about us and how we develop hardware solutions for the world's most ambitious companies, head to synapse.com and then make sure to search for hardware to Save a planet in Apple Podcasts, Spotify, or anywhere you like to listen. Make sure to click subscribe so you, you don't miss any future episodes. On behalf of the team here at synapse, thanks for listening.