 Good morning, everyone. And for those of us in the audience that are from further away, good afternoon and good evening, my name is Jimmy Chen. I'm the managing director of StorageX. And I'm here today to host a very exciting symposia, establishing battery production in the US, stories from the front lines. And we have two amazing speakers that will speak to us. This is Abraham Anapauski of TRI, and also Kurt Kauti of Scylla Nanotechnologies. So with that, I'd like to start with Abraham, and let me do a quick introduction. So Abraham Anapauski, Dr. Abraham Anapauski is the director of advanced manufacturing research at the Toyota Research Institute, TRI in Los Altos, California. He has worked in R&D manufacturing and semiconductors and clean energy for over two decades. His work has focused on materials discovery, device optimization and data-driven methods and batteries, photovoltaics, magnetics, and a variety of other material systems. He obtained his BS in physics from San Francisco State University, and his MS and PhD in Material Science and Engineering from the University of California, Berkeley. With that, I'd like to welcome Abraham and turn it over to you. Thank you very much, Jimmy. And it's a real pleasure to be here speaking at this symposium. And I'm gonna give you a little more background about what I did because I'm joining very illustrious speakers in the past, like Jeff Don, and I'm not quite as well known as Jeff Don. So my talk today is going to focus on the need for collaboration through emerging capabilities and informatics and U.S. manufacturer as a way of ensuring that the path towards vehicle electrification runs smoothly. So as Jimmy mentioned, I've been working in clean energy for a long time. I, at Berkeley, I worked on capital discovery and synthesis with Mark Adoff and other people and since then I've worked in a wide variety of technical engineering efforts in lithium-ion batteries. In the last six years, I've been really lucky to be a Toyota Research Institute working on data-driven methods, first in materials discovery and battery informatics and recently over the last three years in manufacturing. And in my career up to the point where I started working on manufacturing, doing the engineering and science of batteries is a different undertaking. There's a lot of art in manufacturing. There's a lot of constraints that you don't have when you're focused on engineering or optimization of a process. And so getting new technology or new methodologies into battery manufacturing is much harder than it would seem as many people here have probably experienced developing a new product or a new technique in lithium-ion battery manufacturing. It's very challenging. But I wanna make the case that there's a way to facilitate that and tools. So the reason why is because we need to make sure that in the process of building electric vehicles that we are able to let everybody in society into the game. Right now, electric vehicles are very expensive relative to the price of any other car. And we need to make significant improvements in the price, safety and reliability of lithium-ion batteries. Traditionally, lithium-ion batteries, the usage case kind of allowed for inefficient production, let's say, or lack of kind of improvement at a sufficient rate. But recently the market and the application from lithium-ion batteries has changed from consumer electronics to vehicles. And that introduces all these new considerations that we have to deal with. But at the same time, over the last five years, new methods for data-driven approach to understand batteries and also now understand manufacturing have kind of come online starting from academia. And I think the reason we can work cooperatively is that the market potential is huge. It's a rising tide. So the traditional model, kind of a low margin, zero sum game and competition is no longer true in the US. The other thing about this is that lithium-ion battery manufacturing traditionally has involved a lot of kind of inside know-how, people, skill and experience. And as we expand the number of, rapidly expand the number of manufacturing plants, manufacturing capability, those few people who have that experience are going to be sparsely kind of distributed. And so we need other ways to share knowledge and allow for efficient and effective battery manufacturing. So just to illustrate this point of getting, letting everyone have access, so that I compiled this list of range and cost for not all EVs, but the bulk of the ones that are out there on the road. And the average price is about $63,000. Right now the average price of a car in the US is $48,000. And that is an all-time high. So relative to inflation, the price is expected to go down. There's a sort of scarcity that's driving the price up. So that difference is actually probably greater than it seems. And why is that? Oops, there we go. So traditionally, the lithium-ion battery manufacturing culture is a low-margin business. And because of the low-margin aspect of it, people are generally risk averse. So change, even if it's positioned as a good change, is an unknown and it's a risk. So generally the type of change that's available or that's embraced is relatively kind of low. And the real emphasis is making sure your battery factory runs kind of status quo. As I mentioned previously, it's a zero-sum game in terms of competition. It's not an expanding market or the market is expanding very kind of slowly with personal electronics. Cell phones drove that, but it's not driven to the point where people would undertake significant kind of reformation or innovation. And the other thing is there's no real pedagogy. There's no shared knowledge. So you have each one of these manufacturers kind of closely guarding their secrets. Now, I think this is the thing in my years of battery work that's always been frustrating is I know there are a lot of experts out there in the audience, people who know more than me, but I also know there's probably a few things I know that they don't know. And I think driving the industry to a better place requires as much sharing of information as is reasonably possible. But it's, you know, part of the reason for batteries being lithium-ion batteries being difficult to manufacture, they're very complex systems. It's hard, you can't take them apart and see like, oh yeah, this thing right here didn't work or the tab is slightly misaligned. Once they're assembled, that's it. They go out the door, they either work or they don't. But nonetheless, to the credit of people working in the industry, the energy density has increased significant amount in 30 years. One of the ways to see that is the energy density of what the 18650 format has almost tripled. And so now you can regularly buy 18650s, for example, that have three and a half amp-power capacity and pretty good rate performance and good cycle life. A lot of that improvement actually though has been in the focused on the inactive components, the separators, the current collectors and so forth. Materials innovation, like moving from LCO to NMC and NCA have contributed somewhat, but I think one of the paradoxical examples is LFP is lower energy density than LCO. And yet LFP batteries have higher density than the initial batteries from Sony. So the low margin. So on this plot on the left, this is the 2022 profit margins for a variety of battery manufacturers given as revenue versus net profits. And then plotted along these diagonal dashed lines these are the limits of different margins. And you can see most manufacturers fall in the kind of one to 5% regime. The vertical lines indicate for given revenue how, where you hit the net profit kind of margin line and the kind of solid light coded lines are historical averages. And so, you have this low margin business and at the same time, you have a situation where you have relatively low yields. Now, I put low in single quotes because it's actually very difficult to find information about what for a particular factory or a particular manufacturer, what their actual yields are. And some of that depends on how you define yield because yield could be, from the minute you assemble the cell that number divide or that number compared to the number of cells that go out the door as validated cells could be the total material inactive materials, active materials and components that go in versus those that go out. Some people would just argue that it's your net operating margin. But however you slice it, it's not good. I think the number that a lot of people agree on is about 80% although different manufacturers can be higher or lower than that. And certainly during the startup phase of a factory run for 15 years, about a year that is spent at extremely low yields in the kind of 50% or less regime. And if you amortize that over time, it doesn't look that great. So, the existing paradigm that has allowed this to kind of go through is that when you look at safety events traditionally, I mean, it's not good when your laptop catches fire, especially if you're on a plane or it's in your house and it burns the house down or your phone catches on fire. But there has been relatively few fatalities or serious injuries due to this. And in addition, if it's just a quality issue, then the replacement cost for something like for conventional lithium-ion battery applications on the order of $10. Now, the paradigm moving forward from in the 2020s is that there are extreme safety consequences. So, for example, this is the Felicity ACE, this famous case, I believe the cause was a Porsche take-in caught fire and the vessel was loaded with other EVs, lots of the VW EVs and they all caught fire and they couldn't put this fire out and there was significant kind of destruction. And then on top of that, the consequence of even if you say like one bad apple, potentially a cell that catches fire or is damaged, potential replacement costs are in the kind of $10,000 regime. So, both the safety, the consequence of quality control and then the potential costs are very, very high relative to the first 30 years of lithium-ion battery production. And with this new paradigm, there are even more worries over margin because, excuse me, the profit margin and automotive is already low. Toyota is consistently one of the best performers when it comes to profit margin, but that's 7%. So, if you add these very expensive packs into the car, you really need to make sure that you can control that all the way through the value chain. And that starts with making efficient use of materials and being able to make good predictions about and good quality control about how the cell will perform in the field. I think the other kind of interesting thing on here is that Tesla, which is, you know, they're stated numbers, they're very profitable now, but if you amortize it over the last five years because the investment in battery production, it's significantly impacted their profits. Now, I think they're gonna go on and be, you know, continue to be a profitable company. But it does point out the fact that when you make these initial investments, Toyota's stated investment right now is around $6 billion in the battery factory in North Carolina, you know, it's gonna eat into your profit margins for a time to come. And we wanna make sure that when we go into this, this is really a change in the vehicle space and not just a kind of both fad. So, the good news though is that while in historically, the nature of battery manufacturing is really contained in kind of the experience that people on the line, there is an emerging field of data-driven understanding of batteries and in fact, an emerging field of informatics and data-driven manufacturing. But we need better yields. We need better diagnostic of potential safety issues which I'll talk about in a little bit. We need better prediction of performance. And we need personally for me, I think one of the things I would love to see is the ability of a line to rapidly adapt to new materials, new opportunities. I wanna emphasize again that I'm talking about the manufacturing space within the US. How do we make US manufacturing more competitive because the IRA legislation is going to mean that to get the rebates, there's all these rules around sourcing and it's not immediately clear that we're gonna have sufficient material supply and whatever we do have you wanna make the most of and in addition, any innovations in the material space we would like to be able to capture rapidly. So, here is a little blurb from SAP and there's a lot of Siemens and all of these folks have these kind of industry 4.0 or factory in the future, depends on how people call it different things. And while they look great and I've had a lot of experience with informatics platforms and manufacturing, actually the reality is each individual application and manufacturing is special and these things have to be fine tuned. So this is not a panacea and in particular battery manufacturing is extra special and what we have found is that it actually requires a tremendous amount of human work, problem solving, really trying to understand the process so that when you use data-driven methods to optimize the process, you're getting value out of the undertaking out of the investment to go to data-driven methods. So I'm just gonna touch on this a little bit and I wanna stop here and point something out. The irony of this talk is I'm making the point that we should use data-driven methods and we should share data with, we should share data with each other but I think some of the most compelling argument that I can make for sharing data is based on my experience with some of the work we've done with battery manufacturing and I can't share that. So I hope you appreciate the irony. I would love to share some of the results here but I can't and I hope to change that situation but let me just say that one of the main challenges is that battery manufacturing is made up of hundreds of steps where data is recorded and thousands of different type of data for every cell that goes out the line there's tons of cells. So it's a lot of data and we really have, as my colleague Matt Gordon says, we really have to have all the data and at TRI we're working on this problem with our partners throughout Toyota and we're often asked what data do you need and we always say we need it all. The additional challenge is that the data types can vary so everything from physical properties like the way to electrolyte to your time series formation data, images, all sorts of things and really integrating those together and having a good way of interacting with them is very challenging. So taking an off-the-shelf product really actually by the time you're done that off-the-shelf product is gonna look nothing like how it started and if it doesn't have that flexibility it's not gonna be very effective. So the work actually involved in doing this is much greater than simply purchasing a system kind of off-the-shelf system. I'm having some problems with my screen. So in our experience in working on this in the real world there is a just as I said I wanna emphasize there's a lot of human effort. The highest priority right now is getting people into the kind of data mindset that we have to capture all this data and make it accessible. The other thing we've seen is that battery production the issues that are happening now are not the issues that are gonna happen in two months. So there's very dynamic shifting kind of problem set. And so using things like machine learning in this environment even to get a good simple model to work in a multi-physics environment is very challenging. It's powerful but again it still requires really good engineering and problem solving. Machine learning is a tool, it's useful under certain circumstances and so you have to determine that. The other thing that's very challenging is getting an informatic system or data-driven tools to be utilized not only to be useful but to be utilized and that's really a people problem and that involves training, it involves different ways to motivate people to use it. Now I happen to know through my underground network that there are battery manufacturing companies where people in the US, this is two people that one that worked with me and one that worked in one of our research programs where they are using this approach. And so definitely people use it. I don't know if all the larger OEMs are using it but certainly some of the newer ones are using it. So I know firsthand from my experience that this works and from other people's experience that this works. It works with sufficient data. So let me jump ahead to the end here and say, I think one of the areas where the battery manufacturing industry in the US is going to make a very strong case for working collaboratively is around the issue of safety. And I think this because, if you have an EV, have a battery fire or some safety issue because of batteries, I think it kind of tarnishes the overall EV industry. I don't think you think people distinguish that, oh, that's only this company's problem. It's not, for those that you remember it's not like the gas tank being in the rear of the pinto and exploding on a rear impact. I think this is kind of, it would be perceived as an overall problem. In addition, I think tackling this together helps all of us at the same time. So again, my colleague, Matt Gordon is giving a talk at AABC in Europe in a couple of weeks. And we're going to explore this idea of getting people to be interested in figuring out really effective, cost effective and accurate ways of assessing things, defects in manufacturing. So these large metal particles that can cause shorting, that can cause outright failures in production and then potentially even worse cause problems in the field over time. And part of this is it kind of understanding the process and understanding how you can intervene. And so what we want is we would like to catch all critical defects, defects that will short out the cell. We want to catch all those before we want to catch them on the electrodes before they go into assembly so they can be removed. So you invest minimal cost because that can take out lots of other active material plus the components, plus all the effort that goes into formation and all the steps post assembly. Ideally, when this is where data driven methods come in we'd like to catch the subthreshold defects by taking advantage of some of the emerging work around using the electrical signals from formation to spot these kinds of defects. And of course we would like to be able to integrate any defect analysis with the overall informatics platform to get a root cause analysis for these defects. One of the challenges here is that you have to have very rapid image processing or if you're using an image driven method or you have to have very rapid processing of any method that you're using to detect these because the speed of the line is very fast. Now of course we don't want any critical defects or defects that will become critical escaping the factory. And we really don't want them detected by a catastrophic PAC failure. That's the worst detection of all. So because as I said, I can't really talk about the work we've done in detail. I'm gonna say kind of that I'm hoping to have a discussion because I'm very interested to hear the audience perspective on my opinion and whether that's shared or understood or even imposed. But the one thing I will tell you is that there is a rising tide. The market is expanding tremendously for electric for outtake of lithium-ion battery production for EVs. And that is going to ensure that you don't really need to compete. And in fact, we can collaborate and cooperate because of the market size. So just as a kind of to make the cases of why this is effective, I started my career in a semiconductor and I worked in semiconductor for a long time. In the 80s, the US manufacturer semiconductor manufacturers were really falling behind international competitors. And so DARPA pointed up some money and they sponsored, they formed this consortium called SEMitech and it was industry collaborators coming together to address common problems. And I think, one of the really nice things that came out and I worked on the tail end of this is this process called the copper dual damasing. And IBM was the first one to come out with this process. But getting it into production is a very different sort of beast. And the idea behind this is that you need to replace aluminum interconnects in a greater than 100 nanometer node or feature size semiconductor processing with copper to kind of overcome the electro migration and just the resistive effects of aluminum. However, using copper in silicon is very, very difficult because copper is a triple donor in silicon. So it was verboten in most semiconductor processes when I started working in fabs in the early mid 90s. But IBM came up with some really clever kind of ways to do this and companies like Applied Materials stepped in and worked on this problem with other competitors to really make this a robust process. And this is one of the key things that's enabled Moore's law to progress below half a micron node. So I know it's possible for people to come together. And actually one of the nice stories is this after this was kind of figured out sematech stopped getting government money very quickly and it expanded to actually an international participation. So what started in the US is kind of an innovative way to solve common problems quickly expanded to include all the major players internationally. And I think this is a great example to show that cooperation actually kind of floats all boats here. So speaking of cooperation, I wanna end this talk as I do all my talks with the slide of our ENEM team because this is really where all this innovation and hard work comes from. We have a great team and I would invite anyone that's interested in kind of talking to us about materials informatics or data-driven methods and materials discovery and kind of vehicle electrification to reach out. Okay, thank you. Hey, man, thank you for that. Thank you for that amazing talk. I think there's a lot of very interesting ideas and history that you have included in that. And I'd like to have a bit of a discussion. Maybe we can start with this idea of informatics and sematech. In fact, there's been a lot of parallels that have been made or at least suggested between the ramp up in batteries now and the ramp up in semiconductors that happened in the last century, the latter part of that. And informatics actually in many ways is a example of some of the techniques that were pioneered in that space during that time. I'm interested in your thoughts regarding the parallel given that you were actually involved and part of that ramp up in semiconductors. Any thoughts on that? Yeah, thank you. That's a very good leading question. I appreciate that, Jimmy. So, you know, when I worked at applied materials semiconductor manufacturing does use a lot of informatics. I thought about this a lot. You know, I think semiconductor devices yield themselves well to, they're well behaved, let's say, they're well defined. They yield themselves well to kind of analytical, you know, governing equations, if you like, or analytical kind of relationships between the inputs and outputs. And so there's also a lot of kind of inventory and process step management. So I think informatics is a natural part of that. I think the thing about battery manufacturing is that it does require kind of also deep knowledge, but it also requires a sort of heuristic knowledge. And there's a lot of knob turning and manufacturing. And I think some of the skepticism about using informatics and battery manufacturing is that it's kind of like, well, you're trying to impose this, you know, analytical system on this system which is not particularly well behaved. You know, there's a lot of complexity in it. But I think with kind of, you know, this ability to integrate and to analyze large and disparate data sets with things like machine learning, we've seen that it is definitely applicable in many, many, many cases. As I said before, it's hard. So I really feel that this is a good approach to take. And I think that in terms of asking about the parallels, I think it's gonna be different. It's gonna be harder, but it will be equally rewarding. And, you know, the semiconductor industry really runs on kind of informatics for manufacturing. You know, data is immediately accessible. And like I said, I know that there are manufacturers in the US at different scales that are already doing this, but in kind of a non-systematic way. And I think I could say one thing here today, it's that we really need to get together and establish some, you know, some commonality about how we approach this. I hope that answered your question. It does, you know, and I appreciate you sharing the parallel and the challenges and differences between the ramp up and semiconductors, which were in many ways much simpler system because they're single crystals. And primarily one, you know, one material as well, which you're making changes to. Well, there are some parallels though, because, you know, like modern semiconductors, and I think there's some plot somewhere that shows the number of different elements involved in semiconductors going from like, you know, aluminum, phosphorus, boron, silicon, and arsenic too. You know, like they use tungsten and cobalt and copper and all sorts of stuff in, you know, in the number of elements and semiconductors has expanded. The kind of complexity of the devices is greater than lithium-ion batteries, but the complexity of interactions is significantly less or more controlled, let's say. So it's kind of fantastic. So I'm gonna, I'm gonna highlight some, I mean, the call to create this data sharing community is fantastic. And, you know, and you highlighted some of the challenges with that, you know, and even in sematech in the very early days, I would say that Bob Noyce has written many articles about or often referenced how that was really difficult to bring these potential frenemies or collaborators-competitors together. And some could credit the success of sematech to Bob Noyce's personal connections and his force of personality. In that same parallel, I'm curious, if you have thought about that, which I'm sure you have, but I'm giving you an opportunity to sort of imagine how this would come about in the same sort of environment where you have companies which are investing billions of dollars and asking them to share information that, you know, that they should just keep very close to the chess here. And, you know, with the risk reward that they will get more information to share and therefore everyone benefits and the tie goes up. What are your thoughts on how such a thing could come about and what would be the force of this, the Bob Noyce equivalent or, you know, the team of Bob Noyce equivalent that could make this come about? Well, first of all, I think, you know, maybe this may be the highlight of my career that I mentioned in the same breath as Bob Noyce. And anybody that knows me knows I have very strong personality, but I know Bob Noyce. So I think this is, yeah, it's gonna take a lot of cajoling and convincing. And I did show you the kumbaya picture of sematech, but I know that, of course, you know, being in Silicon Valley, there's intense rivalries. But the deal was that, you know, sematech allowed in large part for at least for American manufacturers and then spreading international, it allowed people to continue following Moore's law. And it allowed the expansion. Actually, the demand was there for increased applications involving semiconductors. You think about the world today versus the world in the 1980s, semiconductors were really, you know, had kind of limited application in the, you know, in the kind of like, you know, commercial space and the industrial space. But that cooperation did in large part contribute to greatly expanded production and consumption of semiconductors. And in the same way, we're projecting and there will be a large market for lithium-ion batteries, much larger than it's ever been, you know, orders of magnitude. So I think there the parallel is exact and the benefit is exact too, right? Because it already costs a lot of money to make a car. And, you know, the battery pack, you know, manufacturers right now, if you want to see what the equivalent of the battery pack is the equivalent of the engine and the gas tank at the same time, you know, manufacturers are that cost for an engine has been driven down. And also the quality and reliability has really, really, you know, highly optimized. So, you know, you're right, getting people who are nominally competitors to cooperate is difficult. But I think step one is to see that we're not, and at least in that space, we're not competitors. That there's a common benefit to all. And then the second thing is to figure out to convince people how to cooperate, right? And that's where I think informatics really comes into play is because if you're already collecting this data, then you have to agree on a set of rules about how to share data. Maybe you have to agree on the human environment, you know, whether it's consortia-based or it's, you know, there's just a repository, how you share results, so on and so forth, yeah. So let's pivot a little bit and talk about Toyota's immense investments now in scaling up of battery production here in the U.S. And I mean, it's clearly something that a lot of companies are now driving toward. And as Toyota is embarking on this, what do you see as some of the biggest challenges that you foresee in this scale-up that Toyota can see? But I would imagine probably similar elements would apply to many of the other companies that are building these battery factories. Yeah, well, I think, you know, there's two, I'm not gonna, I think this is pretty true across the board. Even automotive OEMs that are building battery factories are partnering up with people who've built them before. So for example, Toyota is partnering with Panasonic through PPS, Tesla has also partnered with Panasonic. Some companies are partnering with LG, Chem, you know, so on and so forth. There's not a lot of people that are going at this just, you know, complete greenfield. They've never made batteries before. But having said that, I wanna reiterate a point I raised, which is that traditional battery manufacturing has relied a lot on experience of individuals in the plant. But if you have this many plants and then you go to that many plants, you can't clone those people, right? If experience is the way that you manufacture batteries, it takes time to develop experience. And so that means for a long time, we're not gonna be operating at kind of peak efficiency. And so that's, you know, really to drive the point home, you have to have tools that amplify, you know, smart, hardworking people in the factory. And, you know, Toyota has great, you know, Toyota has the Toyota way, TPS, one of the greatest innovations in manufacturing since the assembly line. And a lot of people use things like lean manufacturing and, you know, Kanban systems. And so, you know, one of the things we really wanna focus on is how to bring that quality, quality approach to manufacturing, the human approach, if you will, and combine that with data-driven methods. And I think one way or another, people will eventually get to that same kind of conclusion that in the lack of having enough people, enough experienced people, they're gonna have to start relying on, you know, new innovative tools to do this. And of course, sharing really lowers that burden. If you only have limited experience and the number of people you need greatly expands, if you can share people's knowledge, you can kind of amplify that experience and knowledge of individuals. How you do it is somewhat up for debate. So to the extent that you can share with us, how is Toyota Planetary incorporating informatics in its factory in the way that you're describing? Well, this is the unfortunate thing is I can't really share with you any of those details, even the ones that I think are fairly easy to get. But I do think Toyota, like other manufacturers in the 21st century is very interested and this is publicly available. They're very interested in this kind of factory of the future or, you know, factory 4.0, if you like. They're receptive to this because it's part of the Toyota way is, you know, Kaizen, it's continuous improvement. And I think, you know, informatics or factory of the future is, excuse me, is the essence of Kaizen. So. You, so you've kind of highlighted that experienced people is one of the big gates. And poking on that a little bit, what about the experience of battery manufacturing line workers or other things, you know, sort of the... Oh, no, that's the experience I'm talking about. You know, you may have, you know, and again, I want to say that I know that there are companies where it's much more of kind of, it's much more people have access to data and share data and learn models, but I think that's the exception rather than the rule. And again, this is because I haven't visited every battery manufacturer and some of this, you have to kind of glean on the download. But, you know, I know it works where it's working, but I think the combination of discipline, sharing and data-driven informatics processes are going to be the key here. Yeah. Okay. Fantastic. Well, Abraham, I want to thank you for your presentation and discussion and we will now turn to Kurt. Thank you. So Kurt Kelty is the vice president of commercialization and battery engineering at Silla where he leads the sales and deployment of Silla's materials. He brings more than 25 years of experience in the battery industry, including 11 years at Tesla, where he most recently served at the senior director of battery technology. He was responsible at Tesla for leading numerous key initiatives, including the companies' battery cell usage strategy, delivering the batteries implemented in the Roadster, Model SX and Model 3, and leading partnerships and material sourcing efforts at the Gigafactory. Prior to Tesla, Kurt was director of business development at Panasonic, where he founded the US Battery R&D Lab. Kurt holds a BA in biology from Swarthmore College and an MS from Stanford and is the author of 16 patents. So Kurt, welcome, and I will turn the stage over to you. Yeah, happy to be here. As you just heard, I'm Kurt Kelty from Silla and gonna talk a little, talk about our company, but just before starting, I've got a one minute video here to share. Hopefully this will come through and then I'll start explaining more about our company. What would you do with more energy? With the power to enhance performance without compromising design? To take the driving experience further, faster than ever imagined. The future of electrification is here and it's powered by Silla. Proven in market and backed by over a decade of research, our next generation battery materials are more powerful, smaller, flexible to meet the demands of the road ahead, enabling you to charge faster, go farther and unlock design without compromise. With Silla, move beyond limitations with the energy and performance to lead the charge and build the future of electric vehicles today. Yeah, so you probably gathered from the video there, but yeah, a little bit about Silla, but let me tell you more. First of all, our mission is really to power the world's transition to clean energy and what we're, yeah, that's our mission. Our vision statement here is to build an enduring company that tackles the toughest battery material challenges with science at scale. And so what we're doing currently is focusing on silicon anode material. So that's what my talk will be focused on today and manufacturing that in the US and some of the challenges that we're facing. But before that, just a little more on the company, we're founded a little over 10 years ago, the technology came out of Georgia Tech and two of the three co-founders came out of Tesla, I also am out of Tesla as well. We do manufacturing in California today and in Washington very soon. We've got roughly 350 employees, they're mostly based in Alameda, where our headquarters is in the San Francisco Bay Area. So what we're making is what we're calling Titan Silicon and it's a nanocomposite material, nanocomposite silicon material. We are the only nanocomposite silicon material that's in market today and you achieve about 20 to 40% increase in energy density, 20% today, 40% in the future, which enables our OEMs to really differentiate from others because they're getting the longer runtime, faster charge and also quicker acceleration from this. It is a drop-in replacement and I'll focus a lot on that because that is just so important here for an industry that is evolving as quickly as batteries are to be able to have a drop-in replacement is critical. There's no new capex required or anything like that. There's no pre-lithiation that's required either. We manufacture our material using only global commodity items so you can procure this from multiple companies, multiple countries around the world and our manufacturing uses just bulk manufacturing techniques so we can scale this up and drive costs down significantly over the next many years. In terms of performance of our material, so today I mentioned that we get a 20% increase in energy density. By the end of the decade, we expect to be achieving about a 40% increase in energy density. So you can see here on the right a chart showing what cells can achieve with graphite. They're basically kind of leveling out in their energy density and then a comparison with graphite with SIOX. That's what Tesla uses now with cells from Panasonic. Show here what solid state can eventually achieve and then showing what you can achieve with Titan silicon material. And then the orange part there is what can be attained by other means combined with the silicon material like for example, improved cathodes. And you can see here the increase in energy density that can be achieved with all that combined together. So that's on the energy density side. The other thing that's really key here is the fast charge where today we're getting about a 20 minute fast charge and when we speak of fast charge that's 10% state of charge to about 80%. And in the future, we expect to get that down to about 10 minutes. And the beauty of this is that you don't get you don't have to compromise anywhere else. You're not compromising in cycle life or safety or in temperature performance or anything like that. We're able to maintain all the other metrics and just improve these two. So where we are right now is we're manually, sorry, we're going to start auto scale production in 2025. And yeah, we'll start in 25 and over the next five years really powering over a million vehicles. So that's what we're aiming for and to try and get into the mass market. So we're starting to the premium market working our way down and eventually hitting the mass market towards the latter part of this decade. We believe strongly that we've got the best anode solution for both luxury and mass platforms. We have a full graphite replacement that's more optimized around luxury or premium. And then we can use a lower blend percentage for more of the mass market as we focus on costs. The other thing that's important here it's a drop in replacement as I mentioned. So overnight you can take a factory and increase the output by 20%. So you can go 10 gigawatt hours to 12 gigawatt hours overnight without any change in your capex using the same equipment, using the same labor. So it really is a huge step for an existing factory. This list are investors. I just wanna call this out mainly because, I mean, in times like we're going through right now where it's a difficult financial environment, fortunately we raised a lot of money in the past and we've got a lot of money in the bank. We've raised almost a billion dollars to date. We're the best financed battery material company ever. And yeah, we're in that fortunate position right now. So let me talk a little bit about performance and expectations in the market. So this chart, it's a benchmark chart showing our Bloomberg chart showing the growth in EVs that are forecasted out through 2040. I think everyone on this call is probably very familiar with that. It's a very steep growth plan. And as far as increased consumer adoption, if you do surveys, you can see that there's a lot of preference for EVs today as the next car for most buyers. And we actually commissioned our own survey. We had a third party do this, a thousand respondents were involved in it. 50% of them were EV owners, 50% were planning to purchase EVs in the next year. And you can see the results we got here which were quite surprising to us. The focus on battery is just really critical here. 75% said they're willing to pay more for a better battery. 89% agree that the better battery technology is something that would be very valuable. And 79% rank the driving range as more important than even fast charge. And we often think fast charge is super important. And it's one of the benefits you get from using our material. But if you ask buyers what's more important, a longer driving range or faster charge, even if you make the difference $5,000, they still opt for the longer range. And if you ask them how much they're willing to pay, we were shocked by how much they're willing to pay for that expert range, roughly 12 to $13,000. And we've got a detailed, this is on our website, so feel free to go check it out there and all the details to this survey are included there. So let me talk about our material here. So as I mentioned, we've been in business over 10 years now. We've had a lot of material iterations over that time, over 70,000 iterations where we make a material, test it, realize we got to make some changes, we'll go back to do another run. So over the years, we've had a lot of opportunity to test different materials. And as a result of that, we've also been able to file a lot of patents. So we've got over 200 patents or patents pending, covering our material. We really are the sole owners of this nano composite, low swell, silicon material. And so anybody else out there that's planning to introduce a similar product like this will most likely have to license that technology from us. The, I mentioned earlier about replacing graphite completely or also doing what we call partial graphite replacement. So the full graphite replacement is generally where you're trying to really maximize your energy density. So that's one market we're going after. The other thing is you could do with our material is do a partial replacement, where if you're really trying to optimize around some other metrics, including affordability, then you might wanna go with partial graphite replacement. But either one, you can do either one with our material, which we think is unique. We're not aware of any other company that can actually fully replace the graphite. Now, if you look at other companies or other competing technologies, we've kind of taken a chance at listing this up here. So in silicon oxides, those are present I mentioned earlier with the Panasonic and Tesla. Silicon compounds and simple composites is another technology out there, 2D electrodes, novel cell architectures, and then incumbent the graphite. That's why we've listed it up here on the left. And then along the top, what we think are the really important characteristics that you wanna ask one of these, if you're dealing with one of these companies, ask them these questions and find out what boxes they check and what ones they don't. We've kind of broken into performance and scalability. So performance, can they do low blends? Can they do medium-sized blends or can they do a complete replacement of graphite? And as you can see here, like SiOX is in the market today, the silicon oxides, but it's only these low blends, like 3% or 5%. It can't be used in a 30% or 100% because the swell is too great. So that's on the, and then another performance metric is this fast charge. Can you get fast charge out of it? With a high VED, VED is the energy density, the volumetric energy density. So can you get the two of those both together? Obviously very important. And then the other thing is on scalability. Can you readily scale to the auto market? Can it drop in right today, right into the gigafactors? And is it likely to qualify for IRA credits? Some of these like graphite today, roughly 90% of it comes from China. So it's not gonna be eligible for the IRA credits. Now, there are opportunities to manufacture some of these materials in the U.S. And if they do, they will also qualify. But the way it is today, the majority of these materials are manufactured overseas and do not qualify for the IRA credits. Let me talk about CELIS material in market. So this is the Whoop 4.0 and I've actually got one on as well. Many of my colleagues do as well. This was introduced in September, almost two years ago, introduced the market. We've got over a million of these now in the market by Whoop and they made an announcement when they introduced it that this was the first product using CELIS silicon nano material. And with this, by using our material, they're able to get a 17% increase in energy density while still maintaining cycle life and all the other characteristics, including safety that were required for the program. As I say, this has been going on for nearly two years. There's never been a quality issue or anything like that that we're aware of. It's been going very, very smoothly for us over this period of time. Now, that was a partial replacement of graphite. We've also introduced a 100% replacement and this is with another consumer electronics company. They do not make any announcements regarding their supply base so we're not allowed to make an announcement either but it's been in market now for almost a year with this premium consumer device. And yeah, so that's an example of a complete replacement of graphite. We also had an announcement with Mercedes. This was last year, they announced that they're gonna use CELIS material in their G-Class. The otherwise known as a G-Wagon. They announced that this will be the world's highest energy density cell boosting performance by 20 to 40%. You can see the cell level, energy density they expect to achieve from this. And this material will come from our Moses Lake facility which I'll talk about a little bit later in the presentation here. So let me talk a little bit about performance here. I'll give you some data. So we're not a cell maker. We make silicon and a material. However, we do need to make some cells in order to validate that the material is working. So we make in-house is a single layer pouch cell. So as the name says, it's single layer. So it's tiny. It's the cells that we can do internally. But what we do is we work with external partners on all other form factors. So whether it's cylindrical, pouch, prismatic, big cells, some of our, we've had a couple of cells made over 40 amp hours to date. I think the more recent ones are over 100 amp hours as well. So we've had a lot of cells made different form factors. And Joe, you can see here the typical test cell design and it depends on whether we're doing a blend or a full replacement of graphite. You can see the capacity of our material, the anachodian, the first cycle formation. So you can see the relevant data here. And as far as test results, the big thing with silicon material is swell. That's the one reason it hasn't been commercialized to date has always been the swell. And the Titan silicon will, it swells about the same amount as graphite. So graphite generally out to a thousand cycles, it's roughly six, seven, 8% swell, somewhere in that kind of range. And you can see here with our Titan silicon that we're able to keep it below 6% out with grade, with long cycle life. And you can see on the right cycle life that comes with that 20% increase in energy density, you're getting over a thousand cycles in this particular application. The cycle life really is confirmation that we've got swell contained. So these go hand in hand with one another. And then just to show you a slide on fast charge, this is a, you can see different charge rates starting with one C, the green line and then going to 2.8 C with that orange line, that 2.8 C is equivalent to 20 minutes. So that's how we're doing the fast charging. And then on the right, you can see what kind of cycle life we get from it. And again, we're getting over a thousand cycles and with no impact based on the different charge rates. So the material is quite impressive for handling higher energy density and the faster charge. Now, if you look at how this would be, look in a vehicle, the way we talk to our customers is that you can get a 20% increase in energy density or range and over a thousand cycles and less than 30 minute fast charge. So if you put it on a scale here where you're showing how much an advantage it gives the OEM to have that extended range, you stack it up against the average, whether it's the average vehicle or the base pack that's already shipping with that vehicle and it's a huge improvement. And then on the right side, you could see again the fast charge performance, you can stack it up versus competitors and you get a huge reduction in that fast charge which is really appealing to the OEMs. Switching gears a little bit, wanted to talk about, so that's on the performance side and then talking about lifecycle analysis which is a critical element going forward, especially with our European OEMs, they're really focused on this number of what are the CO2 equivalents per kilowatt hour that are emitted from our material. And especially in comparison to graphite. And you can see here, if you compare it with the synthetic graphite versus the natural graphite, you get these massive reductions in CO2 emissions. So, and we're able to achieve this because we've designed the material this way because we have access to hydros, our facility in Moses Lake that I'll talk about shortly is all powered by hydro. And then the other thing is that we get such high capacity. So, when you're doing this on a scale, we're comparing it again with kilowatt hours and CO2 equivalent per kilowatt hour, which is the right metric to use because we get such high capacity of our material, it also enables us to get this low CO2 emission number. Okay, let me talk a little bit about how we deal with customers on the marketplace. So, as I said, we do not make cells, we make the material that goes in cells. And so you would think normally that we would sell to a cell manufacturer. But what we actually do is we try to sell the OEM instead and we work very closely with them to understand what their requirements are because we can tailor not only our particle but we can tailor our recipe because we have an extensive team internally here at CELO where we're developing recipes that have the best electrolyte, the best binder, the best additives so that when we can work with a customer, we give them a whole recipe, not just we're not throwing our material over a wall and saying, here you go, we're working very closely with them. And so we want to work closely with that and we have to understand what their requirements are to really tailor a solution for them. Then we then get introduced to the cell manufacturer by that OEM. We end up doing a lot of work with that cell manufacturer, again with the recipe to make sure that they've got the best recipe to optimize around our material. Now, we're not a cell manufacturer so we're not an expert in making cells but we are an expert in using our material. So we can give them our material with a recipe that works great. They then take that and because they're experts in cell manufacturing, they can tweak that design and really optimize it in a way that we're not capable of doing. So they take it to another level. So we ended up working very closely together the three parties to get an optimized solution. We ended up shipping the material to the cell manufacturer, supporting them or warranties with the cell manufacturer but a lot of our early work is with the OEM. I mentioned earlier about the drop-in replacement capability and this is showing a typical lithium ion factory starting with a mixing. You can see on the bottom left here you got the cathode and the anode separated. It goes from mixing to coating to calendar and slitting to stacking. So that's the typical process in manufacturing. And what we're talking about is just look at that orange area where the anode is the copper color of copper there. And so the only thing we're changing is that mixing area and we're not changing the equipment. All we're doing is changing the input. So you're changing one black powder for another black powder and it goes into the hopper and mixes up and then goes on the line. Now we're changing parameters. So the levers are changed on the equipment and some of the other, what we call co-products are changed as well. For example, the electrolyte will be a different electrolyte. The additive will be a different additive. So you're getting some changes in the inputs but not in the equipment. So that's why you can get this huge increase in output overnight in the same factory. And from this, you can get a great increase in revenue from IRA credits and I'll talk about that later. I'm going to go through this a little bit quickly because I'm running out of time here. We want to increase, when we're talking to sale manufacturers, we increase the factory output, low emissions. It syncs with their supply chain and it's fully compatible with their existing facilities. I mentioned earlier about the IRA. So this is, there's two elements, the IRA, that are relevant with the ANO material. One is this 45X production tax credit. So this is, as many of you know, U.S. companies that qualify, North American companies that qualify, the, actually, I'm sorry, it's U.S. based companies qualify. Get this $45 per kilowatt hour credit for manufacturing when they're in the U.S. And if you then use CELUS material, as I mentioned, you're going to 20% increase in output. So it's essentially an increase in $9 per kilowatt hour right off the top, just by switching to our materials. So it almost pays to use our material. I'm exaggerating a bit, but you get the point that there's a real advantage to using domestically made CELUS silicon material. And then likewise in 30D, the clean vehicle credit. So this is another credit that is available from North American, bonus that come from North America. This is the $7,500 credit. And just to point out that, again, graphite over 90% comes from China today. And so vehicles with this battery active materials sourced from or refined by foreign entities of concern such as China are disqualified from the 30D credits. So again, this is a benefit by having domestic manufacturing using our domestically made ANO material. We manufacture using the highest quality standards. ISO, we've already received ISO certification. We're in processing our IETF certification. We're building just a first rate team here in Alameda that'll eventually transfer up to Moses Lake. Here's our facility in Alameda. You can see the four buildings here. We do are in the second building here that's labeled manufacturing. That's where our powder is made, the silicon ANO material. Then the building on the lower left, that's where we assemble it. We do coatings and we assemble it into single layer pouch cells. The two buildings on the upper part are our office buildings. So it's a nice campus in Alameda, just about 20 minutes away from San Francisco. And then I'm gonna talk about Moses Lake next. So the facility though in Alameda, we've got these three lines that you can see here on the left. So we have our R&D line. That's where we've done our 70,000 iterations. We've got our pilot line and then we've got the Alameda plant line. So that's the Alameda plant line is the one that's producing our material for validation today. And then we've got Moses Lake facility and that's gonna start up. We'll have first powder towards the end of 2024 and then the starter production will be in Q2 of 25. And then we're gonna rapidly expand from there once we get the first line up and running. This is a picture of our Moses Lake facility. So the, we've added drawings. So the, you can see the outdoor equipment in the upper picture. That's all rendering. The building exists is what it looks like today. And then in the bottom picture, you can see the, again, some of these outdoor equipment that those have been added in. Those do not exist today, but the building does exist. It's a 600,000 square foot facility on a bunch of land. Our key raw materials are located nearby the main material right across the street from us. So that's, it's really helpful. It's also all powered by hydro. We've got rail line coming into the facility as well. So it's really a great facility. It's about 10 years old, the building. And it actually never got used. Two different companies owned it, but yeah, we're the first ones that are actually planning to, install our equipment and start manufacturing there. Just the, again, looking at this, so we have two facilities, roughly 150 gigawatt hours worth of production by the end of 2028. And you can see the different facilities and their characteristics. I wanted to add a little bit here on manufacturing domestically and some of the challenges we've faced as manufacturing in the US right now. I think the big thing is there's just a massive amount of factory construction right now. It's $200 billion have been committed domestically over the last year. And that's being driven by IRA as a part of it. And also there's pent up COVID demand. That 200 billion number, that number is normally around 70. And so that's huge demand. And what the impact of that is that it's driving up costs of equipment and driving lead times. So the lead times are longer, which is really, it's painful for us and painful for all the other companies out there that are also trying to scale up right now. And it's causing shortages of contract labor and higher rates for that contract labor. The other thing is that the challenges we have that are more localized is electricity demand is also high. And so it's taking this longer lead time and building factories to make sure you have enough electricity there. We've had good luck with our permitting agencies but state EPA and air permitting are still the long poles out there. So we still have to deal with that. And then there is concern about operational labor shortage in the Moses Lake area. It's not a huge city, but we're partnering with the various schools and trying to recruit and train the local talent there. And then if one of the advantages that we have though, compared to other companies that are trying to do this, we're really fortunate. First of all, we make a powder that's non-toxic. It's got long shelf life, easy to handle, easy to ship. You put them in 500 kilogram super sacks and you can ship them around the world. So it's real easy to ship. So that means we don't have to be located right next to our cell manufacturing partner. We also don't use any raw materials for foreign entities of concern. We don't need that many employees. It's essentially a chemical factory. And so we can make a lot of material with very few employees. So we're lucky that we don't have that huge problem of needing a big labor pool nearby. And then we don't have anything nasty that's being emitted from our factory. So permit process is relatively easy for us. And then as far as usage, there's a big attraction using our material because roughly 90%, as I say, come from China today, the graphite. And we're displacing that. So lots of the cell manufacturers just really want to use our material and they easily want it quickly. And it's doubly important that dropping replacement is super important so that they can just switch over really quickly. So there's no like a long-term switching costs that are involved here, which is a benefit that we have compared to other companies that are entering the battery material supply chain. Summary slide here. Again, we're going into auto scale production 2025. We think we've got the best end of solution for both luxury at that high end, as well as the mass market platform towards the latter part of this decade. 20% more gigawatt hour output really limited development work required for that. So lots of reasons to make that switch over. With that, I'll end it. So thank you for your attention. Thank you very much for that presentation, Kurt. And that summary of both CELIS technologies and the ramp, which is very exciting to see. It must be very exciting for you guys as you have your capacity increased by 100X. One of the things that I have as a beginning question which I thought was really interesting is that you actually work with the OEMs and you fine tune your materials and process to make sure that you create the material that maximizes the OEMs interest. So are you able to... So when I think about that, I think that you might have different formulations or different processes for different OEMs. And if that's accurate, how do you do that within the same factory? Yes, I'm glad you're asking that because I should correct that. So initially, our hope is to really have one particle at the beginning, one material, and then gradually we'll add more as we optimize around it for other customers. But what we're doing is we're using that one material but we're optimizing a recipe around that. So the loading may be different of our animal material. There's a lot of different levers we can pick in the CEL design process which will enable us to optimize around whether it's energy density or cycle life or fast charge. So there's a lot of different levers we can pull on the CEL design side that we do internally at CELA when we're addressing both the OEM and the CEL manufacturer. I see, so currently in your Alameda facility, do you just run one material and one process? Or do you actually play around? Yes, I'm glad you're clarifying that. Yes, it's one process, but it's one particle that we're making right now. Now we've got, as I mentioned, we've got the R&D line. So we're making all sorts of other materials that we are also testing. We're testing with our partners as well. So it's just not in large quantities. But we do make a lot of different particles to figure out what particle is the best that's optimized for our customers. Okay, fantastic. It's very exciting to see sort of a scale of what I call a non-traditional or potential for non-traditional lithium ion batteries, which is what makes this very exciting. As Abraham was just alluding to, scaling up of a battery facility is a big undertaking, which is why in many cases they partner with someone who has that experience or that work experience or that know-how in order to help them scale. So to see a company like CELA which is doing something different and scaling that up is very, very exciting. So with that idea in mind, are you finding that you're able to already have demand for the amount of capacity that you're putting in place? And how are you sizing your factory? And what goes into that decision? Oh boy, that's a good question, Jimmy. So, and it's a good question because it's kind of a chicken and egg where the OEMs, they've got these long development cycles, they're like five year development cycles. They go from A samples, B samples, C samples, each one of these is roughly a year long and it's a long drawn out process. And they don't wanna commit to anything until they've actually seen the final product. So once they see the final material, then they're like, okay, let's start the validation process, but that's way too late. If they wait until they see the final material, that'll be from Moses Lake in 2025. And so then they start a sample so they can deploy the vehicle in 2030. Well, that doesn't make sense. And this is, I think, a difference between some of the OEMs versus others. For example, my experience is that Tesla, Tesla is very quick at adopting something. They're not gonna wait until the final material is out there and validated. And I think what's interesting is that because Tesla's in the market and putting this pressure on everybody else, so Tesla's like, well, you got a great material? Well, I'll test it for a few months. Looks good. Let's put it in the market. I mean, that's the mentality we had at Tesla when I was there. You don't do a five years validation. You do validation to make sure it works and you're validated all along the previous versions. And so when you get several months of data, and it looks very similar to the previous material that had this extensive testing done on it, then you feel really confident that it can go into the market. Now you gotta do your safety testing. You gotta do, there's a certain minimum amount you have to do on that actual material, but the longer term testing, as long as you can say, hey, it's really similar to that previous material. So I'm gonna take that previous material results and apply it here. So that's kind of the mentality they use at Tesla when I was there. And that's putting pressure on the traditional OEMs, which didn't really wanna start that until they had the final material. And so that now they're adjusting. And so when we're working with OEMs, normally it would be, okay, from that A sample, give us final material. Well, why don't we give our current material? It may not be the final material, but let's use our current material. You start the validation process that way. And so how far in that A samples or B samples or C samples, how far in there can you go with your current production that's not the final material before you switch over to that final material? And let's say it's a real challenge for the traditional OEMs to kind of change their way of thinking to be more nimble, to be able to adopt the latest technology. And it's fascinating to see them evolve over time, becoming more and more nimble, more like Tesla. And if they don't, they're gonna get crushed in the market because you can't, I mean, if Tesla's gonna put things in the market a year or two years before others because they're accelerating their test in, boy, they're gonna have a huge advantage over others. And that advantage is there today. And well, I assume it's there because of my past history at Tesla. Yeah, so working this out and getting those qualifications done in time so that when we hit production in 2025 we've actually got customers for that material. And working that schedule out and that including the validation is a real challenge, Jevin, without getting into some confidential discussions. I can say that we're scaling our Moses Lake facility in a way that will meet our customer needs that we've already reached agreement on that they're gonna buy a certain amount of material. So we're scaling as we're scaling appropriately for the customers that we have lined up for Moses Lake. Well, that's a fantastic story. Given part of the way I think of that is the level of risk that companies are comfortable with and based on their exposure they have to make decisions based on that. And it used to be in the automotive industry with the tolerance for risk was very low. So that's why they had such a very rigorous, long periods of adoption. And it's great to see that that is now being changed as these new technologies come in place. And the new technologies are coming fast and furious. So this is, I mean, as far as that, what I can see, this is gonna become more the norm for a certain level of risk tolerance that people are seeing. Okay, fantastic. Let's talk a little bit about supply chain because in many ways you're a critical supplier for a critical component that goes into these batteries. And as you pointed out, the vast majority of the auto materials comes from China. And that's one of the big questions right now is how do we become more independent in that space? So in that vein, one of the questions I have for you is how do you see SELA and what will SELA do to sort of, it's actually confidence that, how do you use that particular aspect in a lot of your discussions? Yeah, so none of our raw materials come from any foreign entities of concern. So that's a real bonus for us. And we're replacing materials that do come from foreign entities of concern. And so when we talk to our customers, it's a no brainer in many respects that they wanna switch to domestically made higher performing material. It's just, it's really, really simple for them. The challenge that we have right now is we're not proven yet. We have not scaled up. Now we've scaled up twice, 100X both times, going from the R&D line to the pilot line and then to our Alameda factory line. So we've done that scale up twice, 100X each time. And we're doing another 100X when going to our facility at Moses Lake. So we've done it twice before, but it's still is a, it's a big jump. And our customers are not completely comfortable until we actually produce that from there. So although they're excited in terms of dropping reliance upon materials coming from those foreign entities of concern, until we actually prove ourselves, it's not that slam dunk that I was originally referencing. You've gotta be able to produce. And then the third pillar there, I mean, we've already shown the performance work. So that's already checked that box. But the one, the other issue is the affordability side. And as I say, we're coming in at that premium end, and then we're gonna gradually work our way down to the mass market, so that we will be cost competitive graphite towards the latter part of the decade. And then at that point, going back to my slam dunk statement, it is a slam dunk at that point. We will have proven ourselves, performance is better than graphite, cost matching it, domestic manufacturing, it's simplified supply chain, lower LCA. I mean, there's all these benefits that the battery pack is smaller, the battery pack is lighter, you get fast charge benefit as well. I mean, there's just no reason to use graphite. It's just a matter of how much material can we make. Okay, fantastic. So one of the things that it's, so coming to some of the audience questions, one of the questions that is from the audience is, how does the consumer electronics qualification differ from the automotive? And I'm gonna ask that, because that is also word for word verbatim from the audience, but I have a follow up question to that after you. So please. So we've, as I mentioned, we got into the whoop device first. And one of the big reasons for that is that the qualification period is much quicker. I mean, I talked about this five year period for autos for consumers roughly a year. And so you can get in there really quickly, you can get in there with a small amount of material, and you can charge a lot of money for it. So for those three reasons, we went with consumer first. It's just, it's so much faster, that validation period. So yeah, that's a nice, nice target for us. But the volumes there, it's not, you can't build a business off of that. It's just the volumes are tiny. I mean, you put all the consumer devices together and we sold everybody. It replaced all the graphite consumer devices. It would still not be as great as one EV program. It's just, yeah. Yeah. The, so that's actually a great example of one of the big challenges that I see right now is that there's a need for fast cycles of learning. And that has to turn over really quick. But there's also a need to demonstrate scalability and product consistency and performance as scalability. And so having to do both is a challenge. And so what I'm hearing from you is that one of the ways that you achieve that is that you have these cycles of learnings for some of your consumer products to hone your formulation, everything else, by which then you can have an established history by which now you can go and go for the more challenging scalability demonstration questions. I wish I could have stated the same way, Jim. That's exactly right. With the amount of learnings, and then we did this purposely because taking something from lab and actually making it into a product that's starting to ship it, you have so much stuff you haven't considered before. Like the consistency, you've got to have, every lot's got to be identical. You've got to have quality programs in place. There's all these things. So that jump from the lab to mass production shipping to customers is a huge jump. And so when we introduced this, it was a big deal. Now, we've got another big deal coming up when we scale up at Moses Lake, but this is, we've gone through one of these big milestones, these big hurdles and we've come out of it looking good. And we've learned a lot from it, as you said, Jim. Lots of learning ahead, but we've accomplished a bunch. Fantastic, Kurt, thank you. Thank you for your presentation. What I'd like to do now is to bring Abraham back. Fantastic. So what I think is, I'll start with, there's a key question, Abraham. In fact, you yourself were describing this, which is one of the key things that you hope for this new factor of the future is the ability to incorporate new materials. That's part of this, or new investments, right? And certainly in the spirit of like Intel and semiconductors, there was this copy exactly, don't change, change control, I mean, which was really very challenging. So for new materials, like what CELIS producing and stuff, how do you see Toyota's factory as you build it, incorporate new advances, changes in materials, and being able to do that efficiently and quickly? So, yes. Thanks, Jim. Good question. I want to thank Kurt for, you know, his talk, dovetailing very nicely and the point I was trying to make about being nimble and adaptive. So, you know, I can't speak for, I'm an employee of Toyota, so I can't speak overall for the policy, but I can reiterate my position from coming from a long background of material science development and lithium-ion batteries, and really, you know, as an engineer, as a working scientist, not understanding why these cool materials we're doing can't immediately be integrated into production. But I do want to say that, I want to echo one of Kurt's comments, that it's a very good strategy to kind of get into a market where there's much more rapid adoption and then use that as a sort of development platform to say, hey, we've validated, you know, scale to a certain extent and we're ready to go in. But I think it goes back to this incredible aversion to risk, especially in automotive battery production because the margins are gonna be much lower and there's a lot more on the line, right? Failures are, you know, when you have 1,000 cells and one of them fails and that takes out, you know, $10,000 pack or $15,000 pack, it's critical. So I get, you know, why automotive OEMs are hesitant to do that. But I think the, again, the emphasis on having new techniques to really mitigate that risk by predictive algorithms about performance or even, you know, monitoring it and continuous prediction in a vehicle, for example, will mitigate that. And all I can say about Toyota is that they are really, one of the reasons that Toyota Research Institute was formed is to try to understand how to use new emerging technologies like machine learning or I guess science really to gain a sort of, to enhance Toyota's already formidable reputation at manufacturing. So for sure, Toyota is open to that. And of course, one of the reasons is, you know, is to be able to adopt new materials like CELA's, you know, CELA's anode product. And definitely people are interested at Toyota and specifically at TM&A and looking at new materials because we acknowledge the supply chain, the supply chain challenges. And we also understand, you know, the reason we're manufacturing batteries is to take advantage of things like IRA and if the materials are produced domestically, that's a big plus, right? Yeah, thank you, Abraham. I mean, it's an exciting and terrifying time right now, at least from my standpoint where I look at, I'm like, wow, we're gonna be having an incredible ramp. I mean, you know, if you just take a look at the projections for vehicles and the necessary supply chain volumes necessary to meet that, not even the raw critical minerals, minerals for that. You know, it's a little bit bewildering on, you know, how are we gonna make all that happen? I, you know, I wanna, yeah, I came in as like, you know, a hotshot battery person. And when you see the challenges actually involved at scale in manufacturing and like building a factory, I've seen pictures, I showed you a picture of the site in North Carolina, I often feel just overwhelmed. Like who am I to kind of, you know, take on this kind of task? But it's not gonna be me. It's gonna be lots of people, you know, doing this, Curt, people like, you know, companies like CELA, all sorts of, we're gonna need all sorts of people and innovative solutions to do this. So I'm just content to try to, push something that I know is effective. But it is, it's really overwhelming. And I do feel overwhelmed at times just because of the nature of the challenge. But, you know, one foot in front of the other and you get through it. Yeah, I actually was gonna ask Curt about that. You were involved with the Gigafactory and Panasonic and getting that off the ground. What's your experience from that? Especially what, you know, what do you bring on that from that when you're talking about scaling CELA? And also a follow up to that is, you're on the front lines of learning on how to scale one factory. But now we're talking potentially many factories. And what do you see as some of the big challenges for all these factories that are in the works? Boy, that's a big question. Yeah, so I, just for background. So I was really instrumental in getting the Gigafactory built in Reno. That was a, in 2013 is when we identified how many vehicles we would be making. We estimated how many Model 3 vehicles we would need, what we were planning to make and how much, how many kilowatt hours, actually gigawatt hours we'd need. And we figured we needed 33 gigawatt hours per year of batteries in order to meet the demand by 2018. We were late. I think it ended up being 2020 or so that we needed that much. But at the time in 2013, that was the total lithium ion production worldwide. And we were forecasting that by 2018 we would need the worldwide production. And Elon looked at that and said that's ridiculous. There's no way that anybody will be able to meet our demands. Cause we had already been working with Panasonic, with Sanyo and with others. And trying to convince them to ramp up was just really, really hard. They were ramping, but when you ramp in the battery industry back then, you ramp you 5% a year, 10% a year, you do something like that. And we're talking about double tripling over a couple of years. And so that's when Elon decided, we got to get in the battery business. And he said, we're gonna make batteries. And at first my reaction was, oh man, what a disaster, what a mistake. There's no way in hell we're gonna make batteries. It just wasn't in Tesla's DNA. I mean, to make batteries, for those on the call that has seen batteries made, it is so hard. It is so hard to make a battery cell in good quality. And you got to do this not just once, not just a hundred times or a thousand times, not even million times, billions of times, you've got to make this exactly the same and you can't make a mistake. And Tesla was still trying to get the alignment right on the front fender and it just, that wasn't what we were known for. And it was chemistry, right? It wasn't mechanical, it's chemistry, totally different. It wasn't electronics, it chemistry. So it wasn't something we were good at. And so I argued back, Elon said, no, we shouldn't do this. And I argued and I lost. And I argued again, my general principle here is that you can argue twice against your boss and then you're gonna have to give up. I argued twice, I lost and I was like, okay, let's try and work within the parameters he gave me. And so that's when I brought Panasonic in. I'm just like, okay. And I had worked previously with Panasonic for 15 years. So I knew them well. And the same guys I used to work for or work together with, they're the ones that led the team. And so we ended up working together at Panasonic and Tesla and ended up building the Gigafactory. Because this building factories is just a huge challenge. And we've got these announcements coming. It's like every other week there's an announcement whether Samsung or LG or whatever is coming up with their factory that they're gonna build in North America. And we don't have the workers for this. We don't have the expertise for this right now. And so it's gonna be really challenging. I mean, what we did at Panasonic in those early days is we ended up bringing a whole bunch of people over from Japan, they had to do this. And I'm sure Toyota, Abraham's team is probably doing the same. They're gonna bring a whole bunch of people over. You would be shocked at how many people we brought over from Panasonic to really help out in that ramp. And they didn't just help out in the construction, in the commissioning, and then in the ramp, and they didn't go home then. There was still, I mean, there's just so many lingering issues that had to be solved over time. And so you need that connection back to your, that parent company that has experienced. So LG and Samsung and Panasonic can build these factories in the States because they're bringing people from Japan. But the rate that this is being announced right now, I don't know how they're gonna do it because, I mean, like Panasonic's talking about three factories in the U.S. and how are they gonna have enough people to do this? And LG has got multiple factories in Samsung as well. So it's gonna be real challenge. I mean, it's one thing to make an announcement, but it's another thing to actually come through and deliver on that. And they're gonna be, each one of these companies is gonna really be stretched because they just don't have enough people that know how to do this. So I mean, I'm optimistic in the sense that we solved it at the Gigafactory or Panasonic solved it at the Gigafactory with a lot of help from Tesla doing that together. But yeah, we're gonna have to replicate that over and over and over again and cross our fingers and hope that we get across the finish line here. But there's a lot of challenges in the industry for the next several years on this. Yeah, and so what you just described is experience from Japan on existing technology. It's not even experienced from Japan on new materials and other things like what Tesla is doing. So in that vein, at the Gigafactory, was there introduction to new materials or how was changes or that, you know, advancements incorporated as part of the operations in a way, because Abraham's referring to that, you're referring to that. You know, we're all talking about control, but yet taking a greater risk as we introduce unknowns or new materials. So how, in your experience at Gigafactory, did you balance that? Yeah, so we had a timeline and working right to left, we knew how much time we had to get that production up and running. And it did not lead to the innovation that we were hoping for. For example, if we had more time, we couldn't develop the equipment. We, being Panasonic and Tesla together, developing improved equipment, whether it's winding machines or electrolyte field stations. If we had a few more months, we could do that. But instead, because we were working against timeline, we ended up limiting the amount of innovation that we did introduce at the Gigafactory. So we ended up copying a lot of what was done in Japan and made some minor improvements on things at the Gigafactory. But since then, I think, I've heard they've added multiple lines and I'm sure they've made huge improvements on those. One of the advantages though of CELIS material is it doesn't require any change. You just use that same line that you were planning to do, use that same line and you can start with graphite. I mean, I think if you're gonna start a new factory, just do it with what's known well. Just start with your graphite, if that's what you're doing. If you're already using CELIS material, then start with that. But start with what you know well and then convert it over. There's nothing, it doesn't require any change in equipment. So that's, again, the beauty of our material. Okay, fantastic, thank you. Abraham, I have a question for you as a veteran of the semiconductor industry. Companies like Applied Materials played a key part, I believe, in this whole ramp up because they ended up doing a lot of the equipment development but also the process development. And they were instrumental in laying, doing a lot of that heavy lifting. In the case of battery manufacturing, is there an equivalent in your mind to like the Applied Material that's, that role, do you see that or do you see that as an opportunity? You know, equipment vendors. So what I know is that in the battery manufacturing space, equipment vendors don't do a lot of process development, at least not that I've seen and maybe Kurt can correct this or back this up. So there isn't that kind of relationship where equipment vendors are integral to production. And for example, Applied Materials will work because they sell to everybody, what they'll do is they'll work, they'll have a marathon, right? Where they'll produce a certain number of wafers with a process to demonstrate to the customer and the customer will provide that spec but it's not necessarily the spec they're gonna go to market with because they wanna kind of keep those separate but it's sufficient to prove it. And I think because the integrated nature of batteries, it's hard to decouple process steps. So the validation really is done on the complete production. And so where you can introduce new materials like a new liner, you know, some variant of tantalum nitride, you can do it at unit process. I think it's much harder to demonstrate unit process in batteries simply because the nature of how things are integrated is much more complicated because of chemistry. You're moving atoms around, not electrons in a battery. Oh, you're moving electrons too but you're moving electrons and atoms. And you have these kind of multiple, your liquid solid interface, solid state diffusion, diffusion in a liquid, temperature effects, all these kinds of things. It's kind of hard to do. However, I think there is an opportunity for equipment makers to make new equipment that solves, for example, like the pre-lithiation problem or improvements in equipment. There isn't a lot of, I wanna reflect something that Kurt said again which is that, yeah, that Panasonic is going to come to the US and help get the factory up and running. But again, they can't just, they don't just check out on the day that the factory is running at whatever nominal production rate or yield it is. They have to stay because that methodology, that school of thought is really, you need that experience. You can't make the handoff until the people that you're supporting have the same level of experience. Now, and again, this is why we wanna institute a data-driven methods right from the beginning because that, we can train people on that fairly rapidly and use their already, people's good intuition and experience with the process and amplify that through kind of informatics. Yeah. Yeah, fantastic, Abraham. Kurt, do you have anything? What is your experience from the Gigafactory regarding equipment makers? Yeah, so it's interesting in the past, Panasonic actually had an own division that they would develop their own equipment for them. So it was all vertically integrated. And over time, what happened is Panasonic, we see Panasonic used to be a dominant force in the battery industry. So they would give big orders to the equipment team, but gradually they became more and more a minor player and these other companies, they were just focused on equipment, could manufacture a lot more, they had bigger volume, bigger throughput and then an exposure to different customers. And so I think they ended up learning faster and driving down costs faster than Panasonic divisions could. And so Panasonic ended up switching their strategy to buying externally, except for one or two, there's two key elements that they would keep in-house in which I can't share, but the other ones they would buy externally. And so by keeping those two in-house, they were able to keep advancing the process development there and work very closely with their internal team on that. But the external one, they would end up work with the external partners, I think there'd probably be less collaboration with them and I'm not in Panasonic now, so I don't know exactly how that's working, but as Abraham said, the process innovation, I think is a bit challenging for the equipment makers unless they're working really closely with that cell maker. And so I'm not really sure how they're doing it today. I mean, vertically integrated has those advantages, but you have the drawbacks of it's generally gonna be more expensive. So do you also share Abraham's assessment that currently the way the process development is difficult to do in a unit because the output really, the only way you're gonna know that it really helped is by the end product, you know, incorporate almost all the whole process. Is that your assessment too that there currently is an inability to do sort of units of that? Well, I guess I take a little issue with that. For example, let's look at the winding machine. Winding machine is just one of the really critical elements of cell manufacturing and there it's making a jelly roll and getting the tension just right and getting the mechanical dimensions all right at the end of it. And it's a matter of throughput. How many PPM are you getting through there? And with those quality standards that is wound correctly, right tension, and the alignment is correct. So you can measure that. So they actually, they can make improvements. It's actually a lot easier to do that thing with batteries. I mean, this is one of the things that just kills me with batteries and we all often get asked in the industry why is innovation takes so long with batteries? Well, it's because to evaluate the batteries take six months and then you build something, you wait six months and it's like, okay, that didn't work. So then you go back and you start over again. But with a winding machine, you can actually see it right there. You're like, okay, did it come through? Was it lined up? Is it, is the tension right? How fast was it? And so you can actually, you can do a lot more, I think. You can make improvements faster. The learning cycle I think is quite a bit faster than it is with batteries. Yeah, it seems like you can actually characterize or have metrics that you can measure for that particular process that you can then make sure that your battery metrics or make sure that your metrics are being met and you can just continue to progress very fast. But if you have to do the whole battery, it's a real struggle, yeah. Yeah, and let me refine that. So it's not that there aren't unit processes as Kurt pointed out. And it's not that in the semiconductor industry, integration doesn't matter at all, of course they do. But I think I should be more careful and say that the balance of, you know, it's easier to do unit process assessment than the semiconductor industry than it is in the battery industry. And obviously, you know, when you have a drop-in solution where it doesn't change anything, that's a preferred kind of space. So yeah, I think that it's more challenging to unit tool development in battery space than it is in semiconductor space. Again, it's not electrons. Yes, no, no, this is very exciting. And it's great to see that, you know, there's just so much factories in the works right now that are being planned or announced. It's a, it'll be an amazing and exciting journey for the next 10 years, I think. We'll all be, we'll either just be witnessing an enormous rapid transformation of our transportation industry. We're going to use to see EVs everywhere, you know, or we're going to be finding that it's going to be really hard to scale and ramp up. And so I think we're all in for a very exciting time. So what I'd like to do is just give you a last few minutes and any thoughts that you would share in general. And then we can go ahead and wrap it up. Kurt, any last thoughts? One point, I just realized that I'm drinking this and it may look like beer, it is east coast here, but I'm not drinking, it is just water, just make sure everyone's aware of that. There you go. Yeah, I'll let Abraham go first on this. Let me think about any closing thoughts here. Yeah, so there was one thing I wanted to say is that, you know, riffing off of what Kurt said about, you know, timelines limiting the introduction of innovation. Like when you have to ramp up, you know, you kind of start to drop all those, you know, grand ambitions and ideas that you had because you got to get the thing, the factory up and running and producing. And especially in Toyota space because our production of EVs and plug-in HEVs and HEVs or eight, you know, hybrid vehicles depends on the output of the factory in North Carolina. However, I think one of the things about informatics approach or data-driven approach is that in the beginning, you can use predictive algorithms and methodologies to kind of say, hey, based on this, I think this is gonna be the output. So you can operate in a sort of shadow mode and you can show people with the data and with the predictions that this actually works. And so you don't, it's not an either or, like you can operate in, you can have an informatics platforms that operates in parallel kind of like I said in that shadow mode until people are convinced that that works. Now, ideally I would like to have it an integral part of production, but you know, the reality is that, you know, as Kurt and I have pointed out that they're gonna bring experienced people into that. And you don't wanna be getting in there, you know, arguing with them and saying, no, no, because you know, you have to get that kind of ramp done. So I think the good news is that data-driven approaches are not incompatible with even a traditional way of starting up. And I think that you can use them right away. Well, it's the choice of the, you know, of the production of who's running the plant to use them. And certainly I think everybody acknowledges this issue of labor shortages and as most people I think are interested in using this approach. So, yeah. Fantastic. Thank you, Abraham. Kurt? Yeah, I guess just an observation, you know, I spent a lot of time traveling, at least half my time traveling meeting with customers, Europe, Asia, and just got back from a trip to Europe last week and I'm going there again next week. And I meet with executives there and it's really encouraging to see what's really changed over the last, I mean, the last 20 years, but really over the last two years or so is just this real change with the execs buying into EVs. There's no longer a sales pitch that's needed. Like you guys really got to be focusing on EVs. That's done. That work is done. The execs now are, it's EVs, this is the direction we're going and they're really clear on that. And now it's just a matter of implementing it. And so it's kind of trickling down with their workforce, like their workforce now is starting to begin to understand, oh, oh yeah. Okay, we're not going to do another generation of the ICE motor, we're kind of winding that down and we're going to instead, we're going to shift a lot of these people that we're doing ICE, now they're going to be over in EVs. And it's really encouraging to see all that activity in the market and we hope to make it better because make it easier for them to make that shift because they, I mean, batteries are, they define so much about the vehicle, whether it's the performance, the safety, the feel for it, the inside, how much space you have inside. I mean, it just defines so much of the vehicle. And so we're hoping to make that easier and make it a better experience for everybody by having longer driving range, faster charge. I mean, you can, if you can charge your vehicle in eight minutes, eight minutes is how long it takes an ICE to charge. If you can charge it similar amount of time, then I mean, there's nothing else. There's no other metric that ICE beats the EV on. Well, maybe top speed, maybe another one, but we don't really care about that. The mass market doesn't care about that. So, you know, I'm really excited about going forward. I think that at CELA, what we're trying to do is make the battery even better so make it easier for consumers to adopt EVs. Okay, well, fantastic. Thank you both for participating in our ramping production in the US, stories from the front lines. It's great to see your thoughts and share your plans with everybody. And just Abraham, I think this whole idea of informatics is one of the key ways that potentially we could scale so quickly is that this becomes something which is widespread, efficiency goes up, it's something which in fact can be deployed and standardized across and shared across the various battery manufacturers. So it's great to see that you're piloting and planning this at Toyota. So with that, I wanna thank you and I'd like to go to the final slide announcing. Yeah, so just join us for our storage tech stocks which are the first Tuesday of every month. These are highlights of energy storage research here at Stanford. And also we have a online course, if you're interested in these various areas, just simply log onto that and join us. So with that, thank you again. Thank you, Abraham. Thank you, Kurt. And we will catch you guys later. Write the best of luck on your ramps. This is an amazing, exciting story to watch. My pleasure. Thank you. Thanks everyone. Bye-bye.