 Good morning, everyone, from Stanford University. My name is Welchu. I'm the faculty director of the Storage X Initiative at Stanford University. It is my great pleasure to welcome you all to today's seminar. The topic today will be a very exciting and relevant one on sodium ion batteries. As many of you can appreciate, the massive scale up on lithium ion battery has created significant supply chain pressure on both raw materials and also active materials. And the fields, both in academia and in industry, have been looking for alternatives for next generation batteries that will overcome these supply chain challenges. And one of the many technologies being pursued is sodium ion batteries. And I'm extremely delighted today to have two industry veterans and also academic experts join me to discuss the present states and also the future of sodium ion batteries. So let me first just briefly introduce the two speakers and I'll dive into a deeper introduction next. So speaking first today with us is Dr. Jerry Barker. He is the co-founder and chief scientist of Faridiyang, a company that is commercializing sodium ion batteries. And our second speaker is professor, Christian Masqualeer, joining us from the University of Piguarty-Jewel-Vernes in France. And he has been working on sodium ion and lithium ion cathode materials for the past 30 years. So I'm really delighted to have both of them with us today to give us a joint view of sodium ion batteries from the industry side and also from the academic side. So first, let me ask Jerry to join us on the screen here. Hello, Jerry. Good afternoon. Hi there, Will. Let me just give a brief introduction of Jerry. He is a true veteran in the lithium ion battery field and also moving forward to next generation technologies. As I mentioned, he is the co-founder and chief scientist at Faridiyang. By prior to that, he has more than 30 years of experience developing lithium ion battery materials from the very beginning. So let me briefly mention his history and also some of the recognitions he has received. For many years, he was the research director Ivalence and one of the original developer of lithium ion phosphate technology that is now widely used for electric vehicles and grid storage. And many of the chemistries he has developed, such as carbothermal reduction for the synthesis of LFP is still widely used in all of today's LP materials being produced. And he has made many contributions to the field, also in academic settings as well and publish many, many papers and extensive patents in the lithium ion battery areas. One, two notable recognitions for his contribution to the commercialization of lithium ion battery is the IBA technology award from the International Battery Association and also receiving the Alessandro Volta Metal from the Electrochemical Society. Jerry, we're so delighted to have you to share your experience with us in the lithium ion battery field and how this could be extended to sodium ion going forward. Jerry, the floor is yours. All right, thanks very much, Will. I'm just gonna share my slides. Thank you. So thanks to Will and Jimmy for this very kind introduction and the invitation to present here. And as Will said, I am co-founder and chief scientist at Faradien Limited. And the title of my talk, you can see it right here is the path to the successful commercialization of sodium ion batteries. A bit of background on Faradien itself, we were founded in 2011. So we were, we think the first company to look at commercializing sodium ion batteries. So it's been a long path already over 10 years. So we set out to develop and develop an IP portfolio and then commercialize sodium ion batteries. And for those of you who don't know much about Faradien, we have two facilities in the UK. So top left hand corner here, you can see the facility that I normally hang out at, which is a shared facility we have at the University of Oxford Science Park, Bedbrook, which is just outside Oxford in the south of England. And top right, this is the other facility we have in the UK. It's in the north of England, it's in Sheffield. And these are the company headquarters. So this is again a shared facility. And right now we make all cell formats for sodium ion batteries at a prototype scale. So we make pouch cells, prismatic cells and cylindricals. Just a bit of news I should bring you up to speed with. At the end of 2021, so kind of about 10 months ago, Faradien was 100% acquired by Reliant New Energy Limited of India. And Reliance has big plans to produce Faradien sodium ion batteries and a new gigafactory in India in the very near future. So that's big news for us. We think that is a big vindication of our technology and a big step forward for Faradien and its sodium ion battery technology. Also, right, some important other news just recently is that IU-PAC recognized sodium ion batteries as one of the top 10 emerging technologies in chemistry for 2022. So we say, we think again, this is an important recognition for the emerging sodium ion battery industry. So right, going back to where we started in 2011, you know, why did we look to commercialize sodium ion batteries? Well, we thought initially this was, you know, we convinced our investors of this is that sodium ion batteries would be an effective and drop-in replacement for lithium ion batteries in many existing applications and also a drop-in replacement for some legacy lead-acid applications as well. At the manufacturing, right, this is an important point, manufacturing, we knew that we could make our sodium ion batteries using the same manufacturing methods and cell formats as those used in lithium ion batteries. And you'll probably know when you're developing a new battery technology from the ground up, you've got to be careful if you develop new manufacturing methods, this is very expensive and very time-consuming. So we thought this was a major advantage for this technology. And we've also maintained, right, that if you develop new gigafactories and the gigafactories that are designed to produce lithium ion batteries, that these can be easily replaced or supplemented with sodium ion batteries. And just to let you know that pretty well all of our companies right now we're helping us out with this technology, make our sodium ion battery cells on existing lithium ion cell manufacturing lines. We have improved sustainability. This is kind of no doubt right. There's no requirement for the use of resource-limited elements, such obviously as lithium, but also as cobalt and copper, you know, in either the cell chemistry or the cell infrastructure. The copper is an important point, you know, obviously you will know that in graphite lithium ion chemistry cells that you use copper as a negative current collector. In our configuration, we use aluminium, aluminum current collectors on both sides of the cell. And there's a great abundance and availability of low-cost sodium salts. And I'll come back to that in a few minutes. All this adds up to a much lower bill of materials. And I mean a lower bill of material, both in terms of the energy market, so dollars per kilowatt hour, but also for the power market in dollars per kilowatt. Safety was an important one. You know, we have an intrinsically safer chemistry than would be available for lithium ion. And also a very low volatility electrolyte. So this ends up giving us improved safety and abuse characteristics. And secondly, in the safety aspect, we have because we're using aluminium, aluminum current collectors both sides of the cell, we can safely store and transport ourselves physically shorted across the terminals. So that means you have extremely safe conditions for transportation and in particular our transportation. So these cells are all physically shorted across the terminals. Supply chain, we say the supply chain right now is generally OK. It's not perfect, but it's OK. And most cell components are available through established lithium ion battery technology supply chains. The electrolyte, we have a bit more because we're using different active materials, we have slightly more latitude in what electrolyte solvents we can use. So that ends up giving us improved temperature performance, both at the low end and the high end. And lastly, we maintain now, right? Well, I tell our investors and reliance and whoever in the industry is, look, you know, sodium ion batteries is going to be a commercial success. It's no matter of question or whether it's going to be a success, it will be. It just now is the question is how much of the market sodium ion batteries will garner from, as I say, lithium ion and some lead acid applications. So just giving you some basics on the chemistry. So around choices, active choices for cathode, you know, for Ferradian, we we use the layered oxide. And schematically, I'm showing you a certain structural type of three layered oxide. But you've also got the possibility of using things called Prussian Blue Analog or Prussian White Analog materials, or just like you can in the lithium space, you can use polyanions like phosphate materials. On the anode, unless you make modifications to the electrolyte, you generally can't use crystalline graphite as the active material on the negative side. So you have to end up using some kind of disordered carbon. And that generally is is a hard carbon material for certain applications. It's also possible again, like lithium ion, that you can use titanate active materials or conversion alloy materials. And these may well end up being second generation materials for here. The electrolyte is analogous to what you'd use for in the lithium space. So it's again, from a commercial standpoint, most companies are using NAPF6, the sodium version of obviously LIPF6 in a mixture of cyclic and aliphatic carbonates. I make the point here that one of those cyclic carbonates that you can use is propylene carbonate. Can't use that in the lithium space because of the exfoliation problem with graphite, but that ends up giving you better. It gives you a lower volatility electrolyte and better liquid range. And just to remind you again that we use we use aluminium, aluminium current collectors on both sides of the cell. You know, what drove us to to to to using sodium initially was this great abundance. You know, we don't need to go into this too much. We kind of know this, right? You compare that the the the earth abundance of sodium over lithium. It's it's, you know, sodium is the sixth most abundant element in the earth's crust. We'll never run out of sodium. You can extract the material easily from obviously from seawater. So we started out with the idea that it would be possible to produce sodium on batteries anywhere in the world using using local materials. And most importantly, right, is we all kind of know this now is in terms of the precursor costs. If we compare the cost was happened over the last maybe five years of lithium carbonate. I think we're all aware of that price increase that lithium carbonate now is running. This is battery grade, high specification lithium carbonate is running somewhere between sixty, seventy thousand, seventy thousand dollars per metric ton. It's extremely expensive. I never would have believed that cost ten years ago, but that's the current cost. You compare that with the cost of equivalent grade, high specification, sodium carbonate and sodium carbonate. It's down at the hundreds of dollars per metric ton. So two or three hundred US dollars per ton. That makes a huge difference in terms of the bill of materials because lithium carbonate or sometimes lithium hydroxide obviously is the precursor material for making a lot of the cathode active materials. That same point goes with what's happening with sodium carbonate in the sodium space that is a massive difference in the precursor costs. And I guess we also know what's happened to the cost of cobalt and nickel over the over the last kind of couple of years as well. Supply chain. I kind of mentioned that the supply chain I thought was kind of reasonable for for most of the materials that we use in a sodium iron cell. Just to remind you right in terms of cathode, you know, we've got this choice of layered oxide, pression blue, pression white, poly anions. And it's fair to say that all the commercial companies really are making these materials either in house or under license. So there is an opportunity for the large lithium iron cathode suppliers to enter this market because there will be a big requirement for this cathode active material as we move forward. In terms of anode active, I said to you that most people are using hard carbon and there are several, you know, two to three, maybe some commercial suppliers of large amounts of hard carbon. And hard carbon was originally manufactured for the lithium iron industry. There is small scale operations for manufacturing of hard carbon in Europe and in the US, but because of the requirements, the performance requirements and morphology requirements are slightly different for for sodium applications over lithium applications. Some of the companies, including Ferrari, and are making proprietary versions of this hard carbon in house. And that gives a benefit in the overall performance. Electrolyte salts really pretty well. Everyone is using NAP of six and there are only two or three global suppliers of NAP of six battery grade in the world. If you use non battery grade and you get all kinds of other issues in terms of insolubles. So I think there's there's an obvious there's an obvious requirement for more companies to enter the supply chain for things like NAP of six. And I'd also add that we need alternative salts to NAP of six. I've got in parentheses there on the left hand side that there is these imide salts, the equivalent to what is available in the lithium space. So TFSI and FSI, you know, why those are in parentheses is because you can't use those generally with the layered oxide materials because at high voltage, the voltage is that these cathode materials operate at. You get problems with the corrosion of the aluminum current collector. So there's a problem there. In terms of the other components in the cell, the electrolyte solvents, additives current collectors, binders and the rest of the infrastructure pretty well all available from current lithium ion suppliers. So that that's good for the industry. Cathode choice is just to give you a summary on these cathode choices, right? So I said there's basically three classes. So the first one, Prussian blue, Prussian white analog materials. Several companies worldwide are looking at this to commercialize natural energy there in California. That's Stanford spin out doing a great job in developing a chemistry on on Prussian blue analog materials, altress in Sweden. And now, you know, a new entrant in the last couple of years has been CATL. We're apparently looking at developing a sodium ion cell chemistry based on on Prussian Prussian blue analog materials. The problem with these for an energy application is that these materials generally have a pretty low tap density. So they've got good power. They've got good lifetimes. If they're based on iron or manganese, they have low cost. But the low tap density and the low true density of these materials generally makes them unsuitable for making high aerial capacity electrodes. So I think they're better suited to power applications, not so suited to high energy applications. The same can be true to some extent with the sodium polyanions and Christian's going to do a great job, I'm sure, in giving you much more information on the polyanion systems. Tia Matt in France is doing a great job in developing the sodium vanadium fluorophosphate material, Na3V2PO42F3, cathodicative material, both Christian and I have looked at for many years. And again, this material is is extremely good for high power applications. It has very good rate capability. But because it's a phosphate, it has relatively low tap density and is not always great for for energy applications. But I'll draw your attention to to Christian's presentation after mine. So really, if you're looking at entering the the energy market for sodium ion batteries, you're pretty well confined to using one of the sodium layered oxides. And of the companies that are trying to commercialize sodium layered oxides, we include obviously Ferradium, Hina battery in China. And now we kind of hear that probably CATL will also be looking at developing a layered oxide chemistry as well. And if you make this these materials correctly with the right substitutions and the right physical properties, then you can end up with materials with relatively high density, high specific energy and very good lifetimes. So all those things together mean that you can, you know, with the high density means you can you can formulate electrodes with high aerial capacity, which are then suitable for for energy applications. I'll draw your attention to the fact also that the sodium layered oxides allows you much more flexibility in the structural types that you can use, much more than you can have in the lithium space. All of the lithium materials that are currently used, NMC and NCA belong to the O3 structural type. But in in sodium layered oxides, because of the change in the the ionic radii of the of the sodium ion, you can also make materials called P2 and P3 materials. And in fact, what what Ferradium has done is taken advantage of this flexibility and formulated many biphasic materials, mixed phase cathodicative materials that give you this benefit of the O3 and the P2 structures. And this going on to that, right, this is a biphasic material that Ferradium is currently using in its energy applications. And until until recently, this material was made by our corporate investor Halder Topso from from Denmark. And for those of you who don't know much about Halder Topso, they are a a very large catalyst company specializing in nickel catalysts. So they have a lot of expertise around the handling of of nickel nickel inorganic materials, both from a chemical and a physical standpoint. And they make in large quantities our next phase biphasic layered oxide cathode material. And in our current design, one of the typical materials we used is shown here and under micrographs, you look at these under micrographs. What you see is they're made from, you know, the glomerates are made from smaller primary particle sizes. The glomerates are somewhere between 10 and 20 microns in diameter. These end up giving you pretty good tap density. If you look under much more high magnification, you get these these intergrowth materials, and this allows you to get the synergistic positive effects of the O3 and the P2. And when we first saw these micrographs, typically the one on the right hand side, you know, we call this one the right, the the burger structure where what we basically have here is is the O3 layered oxide as the kind of bun part of the burger. And then and then the patty being the P2. We also have the reverse situation in this in this micrograph shown in the center. And this particular biphasic material is around about two thirds, oh, three, one third P2. And that gives us this this benefit of of high power capability, but also good high energy density as well. And so to illustrate that, these these voltage profiles here, the central one in B shows the voltage profile in a sodium metal half cell. And you can probably see they're right, we're we're delivering a reversible specific capacity of approaching one hundred and sixty milliamp hours per gram. For those of you who are not kind of sure about what that means, that's kind of a kind of similar figure in terms of milliamp hours per gram specific capacity to what you would get from a lithium ion phosphate cathode material in lithium ion. On the right hand side, we're showing here that same cathode material incorporated in a full sodium ion cell with a hard carbon nano. And even with this configuration was showing at full depth of discharge, a reversible specific capacity of over one hundred and fifty milliamp hours per gram. That's really excellent performance, right? And the operating voltage of that cell, you can probably see the average voltage is about three point one volts for the cell voltage. So so really excellent performance from the cathode active. In terms of the hard carbon materials, I said to you, most of the negative active materials are are based on hard carbon. If you just take a commercial hard carbon, one of the the hard carbons that was originally developed for the lithium ion space, it gives you a pretty good performance in a half cell, right? You get close to 300 milliamp hours per gram. But we've looked at huge numbers of precursors and different synthesis methods to make hard carbons. And if you know that the requirements are slightly different for for sodium applications over lithium applications, you can develop hard carbons that give you superior performance of the commercial versions shown you here, one made by Ferradian. So we're running up at a reversible specific capacity and a half cell of three hundred and thirty milliamp hours per gram with good charge efficiency in the first cycle. Three hundred and thirty milliamp hours per gram is very similar to what you would do, what would get delivered from a from a good crystalline graphite in a lithium ion application. So it's it's good performance. A lecture like that, I think I said to you earlier that this is maybe one of the areas where it needs additional work, right? So everyone in the industry is pretty well looking at the use of NAPF six in a mixture of kind of carbonate based solvents. Well, we talked to our friends at the Faraday Institution. It's a government run initiative in the UK and the University of Cambridge about ways that we might be able to supplement the delivery of NAPF six and Cambridge University did a really nice job in going out and looking at the the synthesis methods that are there in the literature for making NAPF six and concluded quite rightly. I think that pretty well, all of the methods were pretty complicated and expensive to implement. You know, again, they generally require the use of things like HF. The other alternative is to simply sodium ion exchange LIPF six. But that seemed to be a pretty negative way of going because LIPF six is an expensive material. We're maintaining that sodium ion cells should be low cost. If you take LIPF six and add another process that that just makes it even more expensive. So the guys at Cambridge did a great job in developing a new synthesis method by by finding a low cost commercially available pre precursor ammonium P F six and then using sodium metal, in this case, two equivalents to sodium metal over the stoichiometric amount, refluxing, refluxing in in a organic solvent like T H F and having a quantitative delivery of NAPF six in high purity. So this was a really nice piece of work. And, you know, we think this is a scalable process. We took that material up for adding and made those made those to the NAPF six and made that into into electrolyte systems and compared it with electrolytes made with the same solvents, but with commercial NAPF six and basically showed no difference. This right hand side shows the testing of four cells at plus minus C upon C upon five rate and showed no difference in performance or rate capability. So going on to some of our full cell testing. So here we've we've got several companies, our sister companies around the world that are making prototype pouch and prismatic cells based on our cell chemistry. So these are some cycling results from, you know, just pick these out. We've obviously got great many examples, but these are typical results from 10 and power pouch cells running at plus minus C upon five or plus minus C upon three for charge and discharge. And we're running them at different depths of discharge. Four point two to one volt is basically 100 percent depth of discharge. And I'll just draw your attention to the fact that a two things here. One, this cycles with very low capacity fade at 100 percent depth of discharge and that the cathode at these relatively good rates of charge and discharge is still delivering one hundred and forty milliamp hours per gram specific capacity. Reversibly, that's a really good result because we were we were a challenge when we first did this work, most people said, well, one of the problems you guys will have is combining high specific energy performance in your energy products and delivering high cycle life. Well, we kind of knew that if we were going to go heads ahead against things like graphite, lithium, iron, phosphate, we would need both of these things, right? So in terms of specific energy, our rate of progress and development has been pretty good left hand side. I'm showing the rate of progress of in the green line of Ferradian production cells. So right now we're running at about 160 watt hours per kilogram, which is the kind of gray dots on here are comparing that with LFP commercial LFP graphite commercial cells from several sources. So this is this is a very similar performance. And so that's really an extremely good performance in terms of specific energy. And then on the right hand side of this chart, we're showing the cycle life. And here we're charging in discharge of prototype cells and there's production cells running at pretty high rates. And you can probably see here, right, that we're out to four thousand cycles. And that's still above 80 percent of the original discharge capacity. That puts us similar, I think, to to really good graphite LFP cells in the lithium ion space. Big advantage for us, and we claim this when we started this work, was because of the intrinsic safety of the sodium ion chemistry, we should get better safety abuse characteristics. And we've done huge numbers of testing on all kinds of sizes of cell and also using different testing protocols for the safety. So we were showing IEC testing here, but we've also done UL and UN DOT. And we've never had any incident. So just showing on the right hand side, they're right. A crush test on the top and the nail penetration on the bottom. But we've had no problems with any of the tests, including the crush and hot box and nail penetration. Really important for us. And I think a differentiator over any graphite based lithium ion cell is that we can store and transport our cells at zero faults. I said that at the start of the presentation. One of the advantages of having aluminium, aluminium and current collectors both sides of the cell is we can physically short the cells for storage and transportation with no detriment. There's no increase in impedance. There's no detriment in terms of cell performance. And right now, everything that we we we store and transport for good safety reasons is physically shorted across the terminals. So this we think is a major advantage over any lithium ion cell chemistry that uses graphite, at least ones that use a copper current collector. And if you are just drawing attention, I haven't got a lot of time to show you the data, but I'll draw your attention. If you need more information on this, then I refer you to a review article that we wrote about a year ago, where we compared the the shipping of lithium ion and sodium ion cells from a materials chemistry perspective. It's got a lot of good data in there. So I draw your attention to that. You know, one thing we're really pleased about for Adi, and I said we've been going for 11 years, we've been making prototype cells since about 2015. What we show here on the left hand side is we're comparing ourselves with the rate of progress in terms of specific energy for what happened in the lithium ion space. And these are the red dots are what's happened for commercial 18650 type cells. So we all know that, you know, Sony kicked off this this work on lithium ion commercially in 1991 with a with a, you know, low specific energy cell, but that's that was the start point. And you can see the rate of progress that's happened over those intervening years. Well, the green line here down here is what's happened for Faradian. So because we've all come within within Faradian, we've already come come from a lithium ion background. We've used our experience to kind of aid the development and reduce the timeline. So I'm saying right now, you know, we're running at a very accelerated time frame and our current prototype cells are about 160 watt hour per kilogram, but we're predicting confidently that there's no indication that this trend of it of growth will will stop and that we should get to over 200 watt hours per kilogram in the next two years. And I think that's, you know, that's with no change in the cell chemistry, by the way, will be keeping the same chemistry, but just engineering the cell in a more efficient way to get to higher watt hours per kilogram. If we can get over 200 watt hours per kilogram, we move ourselves into a position where we're not far short of the performance, the specific energy performance and energy density performance for NNC and NCA. So we will do that, obviously at lower cost and better safety. And you might say to us, right? Well, that's all well, well, well and good. Like just saying you're going to get to 200 watt hours per kilogram and beyond. But how do you think you might do that? Well, we have a lot of ways we're doing that right now. One of them is very simple, incredibly simple, is that the larger cells you can make. And obviously, because we have better safety characteristics, we can make bigger, larger format cells than is available for lithium ion. So as you increase the cell capacity in the cell energy, you get you also get improvements in the packing efficiency of that cell. So that increases the specific energy in a very simple way. Secondly, electrolyte loading. And generally speaking, we kind of measure that in the lithium ion or sodium ion battery industry in how many grams per amp-power of electrolyte you add. And right now, if you develop a new cell chemistry, this is pretty well maybe one of the last things that you fully develop. But we know that there's a good way for us to go in terms of optimizing the grams per amp-power of electrolyte that we use. And if we get that down to where we need to be, that that has a benefit not just in boosting these specific energy, but also is a very handy cost saving. Because electrolyte now is pretty costly. Simple way again also is increasing the thickness of the of the electrodes. So this is, as I said earlier, getting to high aerial capacity electrodes. So this is generally measured in milliamp-powers per square centimeter. So we're moving that ahead at all times. If you can get to high aerial capacity, then you get better packing efficiency and you need less layers in the cell. Electrode formulation is you get to higher active loadings of the active materials in the electrodes. And we're doing that right now. And as we're showing in this chart, our cathode active is running at 96%. So we push that ahead and we're still pushing that ahead in both cathode and active. And lastly, you have to optimize the electrode density. This is in grams per cubic centimeter generally. And that comes down to how you control the porosity by pressure calendar in the electrodes. You have to get that right also to boost the specific energy. Cost analysis, we keep this under constant review. And we compare ourselves at Faraday and with the best that commercial LFP graphite can do. And obviously this is a moving target. Things are changing all the time. As I said to you earlier, the cost of the lithium precursor is going up is actually working to our advantage. Right now we are suggesting that for energy cells against commercial LFP graphite, we're around about 30% less in bill of materials in terms of dollars per kilowatt hour than the need that the lithium ion phosphate graphite. So that's a very handy saving. Faraday's IP status, I kind of mentioned sitting this in passing at the start. You know, the first few years that we were working in this space, we were kind of basically an IP factory. And we've got IP and know-how that covers everything from the very basic material IP. So the active materials and electrolyte formulations all the way through to BMS and pack design. And I'll just draw your attention to the fact that we also include here that we've got IP covering the zero volt storage and transportation of sodium ion cells, which I think is as you're probably aware a very valuable piece of IP. Who are the world leaders in sodium ion, right? We're not the only ones. I've kind of focused a lot on Faraday and obviously, but there's several other companies out there doing a really great job. And I think we don't consider ourselves to be that, you know, in a competitive environment against these companies. I still think we're all working together to develop the sodium ion market. So Faraday and obviously, you know, we're based in the UK layered upside hard carbon chemistry and we're trying to attack all energy and power markets, right? So renewables, telecoms and even moving forward to battery electric vehicles. That will require us to get to hire what are our per kilogram figures, but that's exactly what we're doing. I mentioned TMAT and I think Christian will mention TMAT as well in his presentation. They're doing a great job. Their gen one chemistry is around power market. It's based on the NVPF material I mentioned earlier. It's a really good from, as I say, power. So power tools and e-buss and what they've announced recently is their gen two chemistry will be around the layered oxide cathode chemistry. So that will allow them to look at energy applications, you know, battery electric vehicles and stationary applications as well. High now energy, high now battery in China have a very similar chemistry to the one that Faraday is developing. So it's around layered oxide and hard carbon. So no big surprise. They're looking at the energy market and some power market as well. Natron energy, you know, good friends of ours. They're the guys in California, the Stanford spin out. You know, I think they're doing an absolutely fantastic job in developing a chemistry where it's all around pressure and blue analog materials. So iron and manganese based pressure and blue analog materials. When you do that, right, you move away from hard carbon on the negative that's going to give you obviously a much lower specific energy. So it's much lower. It's kind of more around the lead acid market, but it has an extraordinarily good cycle life and rate performance. So this means there's really good applications for the Natron energy products around behind the meter application grid. I know data centers is a really big market for them. And just I think two days ago, I saw another announcement from Natron saying that they'd signed an agreement with United Airlines around what they're going to do on the around kind of ground transportation application. So that's really great for them. Ultras, again, good friends of ours in Sweden are developing and doing a really nice job in developing pressure and white analog and iron based pressure and white analog cathode material. If they go ahead and combine that with our carbon, it gives a pretty good performance again around, especially around kind of grid and stationary and large format applications. And lastly, in a one we can't avoid because it's the biggest engine that we've seen in the last couple of years. And it's really kind of buoyed up the sodium market, I think is that CATL announced that they would be entering and actually going into production next year 2023. And they originally announced that this would be around a pressure and white analog and iron based pressure and white analog cathode around a hard carbon negative. And we kind of think this would be maybe complimentary to their lithium iron phosphate products. They've also announced that they might end up using cell packs with a mixture of lithium and sodium iron batteries. That's an interesting application. They're claiming that their specific energy is similar to Ferradians. It ran 160 hours per kilogram. That's yet to be confirmed in products, but we'll see what happens. And we think now that they're looking at layered oxides maybe also maybe a replacement for pressure and white analog to get to the high specific energy. And I'm not bothering to mention there's a lot of other small startups also now available in China and India. So we'll watch what happens there. And just within the last couple of days I see that there's unconfirmed reports that maybe BYD are going to enter the sodium iron space as well. So I'll just leave you with some closing thoughts on where we need to be, I think in sodium iron is, you know, sodium iron batteries will be commercially successful, I think, right? We know that's going to happen now. All we've got to do now is figure out how big the market will be. You know, Ferradian, we believe has a world leading sodium iron battery performance and I've talked about that already. Our specific energy performance is pretty similar to LFP graphite, but at a lower bill of materials. Our technology roadmap, I think without changing the cell chemistry we can target a specific energy of over 201 hours per kilogram. And, you know, hey, next stage we're all thinking about what's going to be the next stage, maybe all solid states, sodium iron batteries, perhaps even anode-free. And I know CATL have announced that that is part of their technology roadmap. Ferradian, we have a world leading IP position. You know, the other commercial companies are also progressing well in this space. We're kind of differentiated in terms of cell chemistry, but I think we're all working together to develop the sodium iron market worldwide. There's a few remaining challenges that need to be looked at. You know, we've got to improve cycle life. We need maybe some more electrolyte choices. We've maybe got to improve active materials. I said about supply chain. Maybe that needs more attention. We've got to push ahead, get some of the bigger companies looking at supplying some of the active materials. Also we need new supply lines for electrolyte salts such as the NAPF-6. But I'll just leave you with the message that, hey, sodium iron batteries are here commercially. We just got to figure out where we go from here. And on that note, I'll thank you and maybe throw it over for some questions. Jerry, thank you so much for that overview of the field and introduction of Faridian's technologies. Really appreciate it. We have time for just a few questions. Maybe let me lead with the high level question first. So you have shown toward the end of your talk that the performance learning curve is very attractive for sodium ion, essentially doubling that of lithium ion. And I think this is extremely exciting. Can you comment a little bit on the prospect of the cost learning curve? For cost, will. For the cost learning curve. So you mentioned about the performance learning curve. How about the cost learning curve? That's going to be an interesting one, right? I think it depends a bit on the cell chemistry. Well, I think in terms of our cell chemistry, I can comment on what we're doing, right? Our cathode material, the biggest cost part, as you'll probably see in that one slide, is very similar to lithium ion, right? It's the active materials are the biggest, are the heaviest hitters in the bill of materials. So it goes in terms of, cathode active is the most costly, then anode active and then the electrolyte. We're very efficient in terms of comparisons with things like NMC and NCA. We're very efficient in terms of how much nickel we have in our cathode formulation. So the B site, if you will, is only one third occupied by nickel. So what we're doing to bring that cost down is trying to be even more efficient with use of things like nickel. Because we know what's happening in the market. I showed that slide where both co-bottom nickel costs have been going up. So we're driving that down. We're looking at being more efficient and even longer term will moving to the holy grail. I think this will be the holy grail of removing nickel completely from the cathode. And so that, but that's very non-trivial because then you end up with having a cathode chemistry that's where the active redox sites are probably only gonna be iron and manganese. I'm sure you're aware of the issues that are associated with using those in the lithium ion space. It's non-trivial, very difficult. I think also the anode right now, I think what we're doing in Ferradian is as we build up the volume of making hard carbon, we're looking at very low cost precursor materials like waste materials, biomass materials for making the hard carbon. The hard carbon, if you buy it commercially right now is a bit more expensive than natural and synthetic graphite in dollars per kilogram. We've got to bring that down as well. Thank you, Jerry. I think this is a very critical point. I think what you have shown today is the performance parity has already been reached between the sodium ion with their layer oxides and versus same lithium iron phosphate graphite. I think that's very much without a doubt. I think now I'm looking forward to when the cost parity would be achieved at scale. And that, I think it's the big question. I think you're right. The graphite is continuing to come down as well and you're intercepting it. And I am confident that the slope will be sharper but nonetheless, I think the big question mark is when the cost. I think you're absolutely right. This is the exciting stuff for us that when reliance acquired Ferradian, one of the big benefits is they have very big plans for taking this to the Gigafactory scale. I think that will be the crucial bit. And I think also we're all intrigued in the industry by what's going to happen with CATL. You know, when they introduce their first product, let's assume it is going to be next year. I think that's going to be, you know, from a price standpoint, it's going to be very interesting how they differentiate between their super-efficient and very low-cost lithium-ion phosphate products. Where's the sodium going to fit in that market? I think that's going to be very interesting. We're assuming it's going to be complimentary. You know, I don't know, I don't know. And they're talking about mixing, you know, sodium and lithium-ion cells in the pack. That's going to be interesting as well. Yeah, this would be very exciting to watch, Eric. I think I myself have been surprised at how quickly LFP has come to become a dominating force just over the past five years. I think something similar may be occurring soon with sodium-ion as well. It'd be very interesting. Yes, sir. Yes, sir, sir, will. Thank you, Jerry. Now I do have one more specific question from the audience. You very nicely showed, I think you termed the hamburger approach. So obviously this is far more than just a blended electrode. It is, yeah. But is there any reaction between the two phases or they are just microstructurally sort of a co-precipitated variance of that? Yeah, it's, what we did originally, well, just put it in the background, we said, well, is an integral structure different just from a physical mixture, right? So we went out and made what we could, it's closer to be good to the equivalent material by physically mixing O3 and P2. And we never got the same performance. So the integrals, just like I think with the mixed-phase materials that you see in the lithium-ion space, you know, the LI2-MM, M&O3 materials, and I think there's a big, there's a synergy from using the integrals structures. So. Do you think that synergy is coming from an interfacial interaction or more of a mechanical interaction? It's tough to know, Will. We don't know at this point. What we do know is from just simple measurements between the two that the physical mixture is not equivalent to what we call the chemical mixture, the integrals phases. So, but it's a very powerful, we didn't go into it in great detail, Will, but you can by altering the synthesis conditions, move from, with the same precursor, right? Move from very high O3 contents to very high P2 contents. So that takes you from basically an energy material to a power material from the same precursor mix. It's a very powerful kind of matrix. I agree. Thank you, Jerry. Sit tight. We'll have Christian present next and then we'll come back for a short discussion afterward. So I can have you join the stage, please. Thank you very much. It is, again, our great pleasure to have Professor Christian Masquelyer give a talk today on sodium ion cathode chemistry coming from the solid state chemistry perspective. And Christian really requires no introduction, especially to us academics, have done pioneering work on poly anion since his postdoctoral work with John Goodenoff some 30 years ago at UT Austin. And since then, Christian has been utilizing powerful probes to understand the synthesis aging pathways of various cathode materials. And we're extremely excited to have him follow Jerry to give a deep dive into the crystal chemistry. I wanna mention that in addition to his expertise in cathodosol estate chemistry, he has also been providing great service to the battery community. He is now the co-director of the Alastor, which is a large European initiative to cross-cut the European Union for battery research. Christian, we're so delighted to have you give a deep dive here on the solid state chemistry, really looking forward to your talk. Thank you very much, Will. And thank you very much. Can you see my slides correctly? Yes, it looks perfect. Thank you. Okay, good. So it's a real pleasure for me to discuss about poly anionic cathodes for sodium ion batteries. Thank you so much for the invitation. And I work in Amiens in France, North of France. And this work is in a very strong collaboration with Bordeaux, the ICMCB in Bordeaux, many colleagues from there. And also for part of the work with NUS, National University of Singapore, as well as the Tiamat Company in France. As Jerry, a very good friend of mine, introduced just before, yes, indeed, there are several interesting families of cathodes for sodium ion batteries, the layered oxides in one part, the Prussian blue, white analogues on the other part, and the phosphate-based sodium ion cathode materials. This is what I'm working on. And you see here are two interesting compositions, and A3, V2, PO4, three times on one side, and A3, V2, PO4, two times F3 on the other side. Just want to point out first that these two compounds, although their chemical compositions look very similar, indeed very, very different in terms of structure, in terms of properties, et cetera. And my talk will be divided in these two structural families of materials that both contain vanadium at three plus oxidation state to start with. If we look at the global picture of sodium ion battery cathodes, there are plenty, plenty, plenty of materials compounds we may think of. And you are all familiar, I guess, with these voltage capacity plots. Just to mention that cation, polyion materials are in this range of energy densities, close to 400 to 500, if we consider the Nasi-Con type in MVP versus the MVPF type of family of materials which show higher voltage when more or less similar global capacity. The layered oxides Jerry talked about have bigger specific capacities and some of them very nice voltage ranges, so which basically leads to higher energy density for the layered oxides. So the menu of today is the following. I will be talking about MVPF first and then I will move on to Nasi-Con type materials. And I like to give somehow the main results I would like to demonstrate in my talk as an introduction. MVPF, as you will see, and Jerry mentioned it, is a very, very high-rate material. Actually, Jerry Barker worked very early on these compounds and even showed it to be used in lithium ion cells quite a bit of time ago. I will show that the crystal chemistry is really fascinating and we discovered a very subtle autonomic distortion that is indeed important to qualify the material and to qualify the stoichiometry. And in particular, as I will demonstrate, the oxygen versus fluorine ratio in this compound that plays a very important role. As you will see, this is a pretty fast ion transport compound and this leads to a very important property of this compound when it is used in a sodium battery, which means the power capability of the cells that will be built on MVPF. We made the operando X-ray under different conditions at different temperatures and at different C rates to show indeed that it has really exceptional properties. And not to mention that around MVPF, there are several patents and licenses that have been taken by Tiamat company in particular with many patents made in France through the RS2 network and with the CEA. So our story with MVPF started close to 10 years ago. Actually, we synthesized this material. It's rather easy to synthesize, I would say, and made many tests with Rosa Palacin from Iqmabe in Barcelona. And we immediately realized the main feature being that it really can sustain very, very high rates at the very early stages of our work on this compound. So this was 10 years ago. And so this led us to really start to work on it deeper and in much details as I will demonstrate. But myself, I'm more of a crystal chemist and I like to look at small little details in the materials we prepare. And MVPF is really fascinating in this sense. First of all, we illustrate here that the compounds when synthesized in a proper way, that means with three chlorine per formula units, is indeed showing a very, very subtle phase transformation, phase distortion, lattice distortion towards an orthorhombic symmetry that can be clearly seen in red here, where we could identify that indeed what had been previously described in the literature was not completely true. And we indeed protected this new orthorhombic phase with a very subtle distortion with as you can see a difference of 1.002 in the B over A ratio, this small orthorhombic distortion. This structure is interesting and this orthorhombic form due to partial ordering on sodium and the crystallographic sites can be disordered by slightly increasing the temperature up to 400K where the transformation from the orthorhombic space group by MMM to the I4 over MMM space group occurs close to 400K above 400K. So this illustrates the very, very fast ion transport of sodium that is completely delocalized in layers perpendicular to the C parameter that allows very, very fast diffusion of sodium into this compound. Second fascinating thing is about the fact that we can somehow tune the electrochemical properties by playing with the fluorine over oxygen ratio into this compound. I mentioned already that the orthorhombic phase of F3 is properly described, but with different synthesis conditions playing with the oxidation states of the vanadium precursors, for instance, we can obtain oxygen substituted fluorinated compounds with a whole range of Y values. And because oxygen actually substitutes fluorine in this crystallographic site on vanadium, so this creates shorter vanadium oxygen bonds, both on the two sides of the two crystallographic sites for vanadium. So basically we can end up with a compound that has one fluorine and two oxygen around the vanadium. This is clearly seen from X-ray diffraction. We see clearly that depending on the amount of oxygen that has been substituted into fluorine, the structure changes and the lattice distortions of the orthorhombic cell changes. And we can really somehow, thanks to precise crystallography and precise X-ray diffraction, monitor and measure somehow the amount of oxygen we have been able to substitute for fluorine. We made a very clear tables of evolutions of the lattice parameters as a function of Y in the fluorinated oxygenated substituted samples. And you can see that as soon as we reach 0.5 oxygen, basically the structure is not orthorhombic anymore and transforms to a tetragonal unit cell. Depending on the amount of fluorine versus oxygen, the compounds show also different thermal behaviors. They have phase transformation from the AEMM orthorhombic space loop to the tetragonal I4MMM. Takes place at slightly different temperatures. So this is also a way to quantify the amount of oxygen in a given cathode, I would say, for this family. And also the energy related to this transformation is depending on the oxygen versus fluorine ratio. As I mentioned, it is possible to prepare a whole series of compounds with different fluorine to oxygen ratio here, F02. So that means that indeed the vanadium oxidation state changes quite a bit. And we might be carefully not actually saying that we have vanadium 4 plus, but we have more of VO2 plus units that are formed when we substitute fluorine by oxygen. You see that basically the electrochemical properties are similar, but they are not in details because you see very characteristics to phase transformations from the stoichiometric compounds with F3. Similar clearly defined to phase transformations occur for the F02 compounds, while we have a much more continuous evolution of the voltage versus composition figures when we have intermediate compositions. This is a general trend. And also what is important is that as the fluorine content increases into this structure, this increases slightly the average operating voltage from 4.04 here to 4.20. And similar for the first intercalation plateau. Then we spent a lot, a lot of time over these at least six, seven years by doing structural studies. In particular, the albacic retron in close to Barcelona in Spain with our colleague François Fort to investigate the mechanism of sodium insertion extraction. It happens that while we mechanism look simple from the voltage composition shapes with two separate features here, we were indeed able to identify many intermediate stages of sodium extraction with many intermediate phases characterized by different X-ray diffraction patterns. So this is rather fascinating. I mean, sodium orders, vanadium charges order as well. Very, very complex actually phase diagram occurs upon extraction of sodium from NA3 here to NA1. This was basically a great work of Matteo Bianchini during his PhD thesis a long time ago. And Thibault Bourg who worked with us for a long time as well. So several features are pretty interesting in terms of crystal chemistry. When we fully extract sodium from NA3 to NA1 here, basically we realize that there are some sub-stall phase transformations that live to a different space loop than the pristine one. And the most interesting feature is that instead of having vanadium 4 plus oxidation states for the given compounds, we realize that at the end of charge for the sodium 1 composition, there is a disproportionation feature that occurs very interestingly. We determined from the X-ray diffraction analysis plus also from X-ray absorption analysis that we don't have a single vanadium 4 plus crystal graphic side, but we have vanadium 3 and vanadium 5 plus coexisting in the same powder, in the same material with these two octahedral that share a corner in the flowing corner. So this is some sort of so-called disproportionation and we were able to identify it for this compound. What we did as well was to investigate the structural properties of this compound and the various conditions in terms of electrochemical cycling. Thanks to the very bright source in the albacic retron and to our operando cells, which performed pretty well, we were able to cycle these compounds up to 25 C in the same retron and identify indeed that the phase transformations still occur even at this very, very high rate and with intermediate phases spotting here. We identified very early, as soon as 2015, that we could try to implement this material in real cells. And this was done in the French task force with Professor Tarasco and Laurent Patricimo, Mathieu Marcret, et cetera. And with the CEA, who was able to build real cells for us with us and to identify that in this structure, this compound was able to very, very stable and in particular under very fast rates. So this generated the creation of the company Tiamat, which is clearly identified now as an interesting player in the world of sodium ion batteries. As we said several times, as Jerry clearly mentioned, at this stage, the NVPF that Tiamat is using clearly makes very, very highly powerful batteries. But I'm happy to announce that Tiamat is also now working on higher energy density materials with layered oxides that will be developed in the near future. Now I will move to the second part of my talk, dedicated to Lacycon type electrodes, based again on vanadium. But I will show you, based also on some extra elements that we try to incorporate into this structure. Around the Lacycon, many of you probably know that it has a very, very high structural stability. This is somehow what I do like a lot within polyanodic materials and phosphates. Their structures are very, very stable for long-term electrochemical cycling. We can play with many chemical substitutions into this framework. So that really allows the voltage monitoring through the inductive effect, as many people know. And this is also quite fascinating in terms of crystal chemistry, where we observe many orders of the phenomena on the sodium sublattice. Not to mention as well that several of these materials have been tested or identified as studied for sodium-soycelle batteries, both for the cathode or not part, but also using Lacycon as the solid electrolyte in between. So yes, indeed, the many groups are tackling problem, tackling studies about sodium ion transport in the Lacycons with different kinds of modeling, of course, different kinds of electrochemical tests with Lacycons, both at the cathode side and the electrolyte. This is clearly a trend. And we see many groups and progress in this field. And of course, we have to acknowledge that the Lacycon structure itself was invented many, many years ago by John Goodinoff when he was looking for solid electrolytes to be used with sodium sulfur batteries. We did a bit of research, quite a bit, actually, on soystate batteries using Lacycon electrodes and cathodes and electrolytes. And quite a bit of time ago also, we were able to build this kind of soystate batteries using NZSP as a sandwich electrolyte assembled by Spark Plasma Centering and to build a symmetrical cell with MVP, both at the anode and the cathode side. The performance were satisfactory, I would say, not to mention here that this where I measured that 200 degrees Celsius, but still feasible. And several groups are working on trying to improve these numbers in solid state batteries that do not contain any sulfur and that are purely made out of oxides. And in this case, with the MVP on both sides, it's a full Lacycon structure and soystate batteries. So what I really like, what I really enjoy, what you have to figure out about the Lacycon structure is that this allows for extensive chemical substitution. We can put basically whatever we want on the octahedral sides here with many, many different transition metals from iron to vanadium, titanium, niobium, manganese, chromium. And then, depending on the oxidation state of these elements and depending also on the anion here, I mentioned only formulas with PO4, but PO4 can also be SO4, can be SiO4. So then depending on these numbers, we can adjust the number of sodium ions into the cathodes. So we have basically electrodes where there is no sodium and BTI is a good example. And we can end up with cathodes also with sodium ions. So really a very nice playground to identify different kinds of electrodes. We can play with different redox couples that operate on these materials. So different redox couples will build electrodes that will function at different voltage versus sodium. So this is somehow very useful for monitoring the voltage you will obtain and also to create, as it has been done by some companies in the US, to eventually envisage actual batteries, sodium batteries based on Nasiqon electrodes. Many, many people worked all over the world and for the last 30 or even 40 years on many different electrodes. This somehow, this graph somehow gathers many of the interesting materials that were investigated from iron sulfate here, as you can see, that was investigated by Monteram in particular from the NATI-2PO4 three times that was discovered by Claude Delmas a long time ago from the chromium phosphate, the highest voltage obtained in the Nasiqon framework by Atsuo Yamada. And also a lot of work had been done quite a while ago also by Palani Balaya in Singapore showing the very high rate capabilities of this MVP type electrodes based on the Nasiqon structure. So we played quite a bit with many of these compounds and depending on the transition metal element you put with aluminum, with Vassus vanadium, I would say, we can have many different properties, many different operating redox voltages. We can substitute vanadium by aluminum here. We can substitute vanadium with titanium with iron, as I will show a little bit later, and with manganese. In these two cases here, because we can substitute with iron 2+, and manganese 2+, we can end up with a cathode containing four sodium ions, which is indeed very important if we want to increase the energy density and the capacity of these Nasiqon electrodes. I will focus my next 15 minutes on the MVP system, and just to illustrate what we recently found on this very highly investigated system in the world. That looks very simple. MVP here, and that's what people prepare, can be either oxidized by sodium extraction at around 3.4 volts versus sodium towards anyone. But it can also be used reversibly as a negative electrode at working at 1.6 volts versus sodium. So somehow this shows you the feasibility of operating on more than one electron per vanadium from NA1 to NA4 in a single MVP electrode that can play at both sides, a negative or a positive side for sodium ion battery. So this system apparently very simple, gave us some very nice surprise recently. Gave us, but also some to some others, we have to acknowledge that some people already identified that in this apparently boring redox plateau here, some intermediate phases would appear, you see, somehow at the middle of the extraction of sodium from NA3. This was even more clearly identified and described by our colleagues from Russia in the Antipov group, which from NA3 here to NA1 here identified the appearance of a new phase at mid charge. So this was really intriguing for us. And we worked a lot at the albacic electron to try to identify this weird intermediate phase. And the slow rate at equilibrium condition, we go basically from the NA3 to the NA1 phase through a two phase reaction that appears at 3.4 volt basis sodium. But if the battery cycle that higher rate and in the different conditions, we indeed also saw the appearance of this NA2 phase at mid charge. So this was really a surprise for us and confirmation from what had been published by the Russian group. And then we undertook many other studies, with upon cycling at different rates and for extensive cycling up to six cycle here, where we always saw the existence of this intermediate phase between the NA3 and the NA1. And definitely it is there, it is there. And we could by carefully monitoring our electrochemistry, we could indeed isolate this phase, isolate where you can see here that the diffraction peaks of this NA2 phase that basically had never been really fully observed before and determined, occurs in between the NA3 pristine one and the NA1 pristine oxidized one. So from the very high quality of this synchrotron X-ray diffraction pattern, we were able to determine the crystal structure of this new NA2-V2-PO4-3 time phase. It was a little bit challenging and my colleague, Jean-Noël Chautat, with Sunkyu Parker, our student and our collaborators, was very hard in trying to figure out the real crystal structure observed. We ended up with two models that we couldn't really decide on to which was the exact right one, with two models where one of them showing sodium vacancies in one crystallographic site and when the other model shows a full disorder of sodium anti-crystallographic sites while the two specific lantern units of the Nashicons framework are oxidized to Vanilla 4 plus on one side and to Vanilla 3 plus on the other side. So this is pretty fascinating to see that in apparent very flat voltage plateau, we indeed have a crystallized well-defined phase just in the middle, in between the NA3 and NA1 phase. Many people, I think, have been trying to get further in MVP and to try to fully extract the third soldier. I was just talking about NA3 to NA1 here that gives an operating of voltage of 3.4 but for capacity is still limited to around 110, 120 mA per gram. The ideal case will be to be able to reversibly extract the third soldier towards V2P4 three times which to my knowledge, nobody really succeeded in doing recently. This is a figure taken from this paper published in 2014, but which is about modeling and indeed the operating voltage will be 4.6 volts but we never succeeded in getting V2P4 three times electrochemically. Very recently, we made a nice discovery, I think, together with my colleagues, which we patented and that I'm happy to disclose today. It's that indeed the conventional NA3 V2P4 three times works at 3.4 volts versus sodium. And we were able by some tricks that I can completely reveal today to increase substantially the operating voltage of this nasi-con electrode towards a 3.6 volt average versus sodium. Besides the intermediate phases that I demonstrated earlier, here we have a full solid solution from the beginning to the end with two waves, okay, but a solid solution occurring. And interestingly as well is that this new material here, new composition leads to a smaller expansion and construction upon electrochemical cycling. So this is pretty interesting in terms of applications. I will move on to the last small part of my talk towards the new studies we've recently performed around the substitution of one vanadium by one iron to lead to the NA4 FEV P4 three time cathodes. Several groups, including coworkers which turned good enough have been working on substituting vanadium by different elements such as manganese to lead to this very impressive electrochemical properties when the operating voltage is limited to 3.7 volts versus sodium. And we indeed also worked on that together with some group in Russia as well to demonstrate that three sodium could be indeed extracted from this NA-MNVP4 three time material. So this leads to a substantial increase of the capacity but this also leads to a substantial change in the electrochemical voltage composition data and curves and shapes with a very strong structural rearrangement upon the last sodium extraction phenomenon. We besides this manganese substitution we worked actively on the iron type some compounds and we first prepared this new compound NA3 FEV P4 three time. So it's rather easy to prepare but the crystal chemistry is not that trivial and it shows as many sodium three contained materials. Very nice sodium and vacancy ordering that leads to very complicated superstructure with a huge unit cell and monoclinic distortion. So we prepared this material easily and we tried as well then to prepare the NA4 FEV compounds with reduced iron which we never succeeded to do by simple soy state chemistry. We succeeded however to prepare NA4 FEV formula by chemical sodium of the NA3 FEV that we had prepared before. So indeed we were able to obtain it as a very pure phase with all the sodium sites completely fully occupied normally because we have all the NE1 and NE2 crystallographic sites occupied in this compound and we were able to confirm as well from Mozboa spectroscopy that indeed iron was reduced at the two plus oxidation state into this material. So we have a compound NA4 FEV that is obtained electrochemically and chemically by sodium sodium by chemical sodium and we were able to indeed determined that the environment around iron, the environment around vanadium looked very similar from the X-ray point of view but we were able to separate the respective contributions from exhausts and determined that indeed we had the iron two plus and vanadium three plus in the same cathode. The electrochemistry is pretty fascinating. It shows three extraction phenomena up to 4.4 volts per sodium with the progressive oxidation of the manganese of vanadium and iron and you can see that basically the phenomenon is not completely reversible as we have only two phenomenon in the subsequent discharge. This shows a rather high structural stability of a recycling with many intermediate phases occurring upon sodium extraction insertion and we were able thanks to X-ray absorption spectroscopy thanks to Mozboa spectroscopy to fully determine the mechanism of sodium extraction from these compounds. At the end of charge, very interestingly we were able to identify again very short vanadium-oxygen vanadyl-type bonds that occur at the end of charge and that are mostly responsible for the structural irreversibility upon the next discharge. So I will conclude about the nasi-con phases just to remind everyone that indeed this structural family that has been identified for very fast iron transport as a solid electrolyte is also a very good fast rate material for cathodes in the sodium ion batteries. The crystal structure allows for tuning the operating voltage towards the application we may wish and many groups now have been focusing on trying to extract three electrons per formula unit in particular playing with manganese and vanadium and this is not completely solved in terms of structural reversibility and long-term cycling but this is one path to take on these materials. I will end up by thanking my good friends and the strong collaborators in Bordeaux, in particular Inamia, of course, my colleagues Vassal and Genouel and in the different facilities such as the Alba-Saint-Côte-Ren with François and much of this work has been performed by Sunker Park, by long MBN and Matheur Bianchini who are no longer with us because they are doing a great job elsewhere now. So I will thank you for your attention and I'm sorry again for this interruption of my internet access. Good job, thank you very much for that deep dive into the sodium ion cathode and the crystal chemistry. So we have time for a few questions from the audience. So maybe let me get started. The first question comes from the use of two electron and three electron redox in many of the poly anions that you showed which I think shows the versatility and stability of the poly anion structure. The one question from the audience is how does the electronic transport alter as you go through these many redox in the transition metals? It basically doesn't. Very, very, very early. We don't have much knowledge about that because basically nobody really took care in precisely measuring the electronic transport but we don't see any difference and there should not really be any difference. The octahedral are all isolated from each other. Really, and they are completely. There is no corner age or phase sharing between any of the transition metal octahedral. So that's a strong point to just figure out that basically there is a super small electron transport. The transport is mostly ionic. So this is a difficulty, right? In terms of the, that's for the cathode but carbon coating, ball milling, advanced coatings on this is sufficient to bring the electrons but intrinsically in the structure of the nasiqons electron transport is not modified by having different elements. I think this is very interesting. I think another way of saying it is that the electronic transport is generally very poor in these materials and through extrinsic methods such as carbon coating you're able to mitigate it. But nonetheless, I still find it quite fascinating that the MVPF has such incredible rate capability despite some of these shortcomings on the intrinsic electronic conductivity of the material. Do you think this is resulting from? For the MVPF it is slightly different. You have shared corners given by octahedral units but the electronic connectivity is not much bigger than for the nasiqons. Yeah, really the sodium, you know, the sodium takes it all in terms of it moves so fast and then somehow I don't know if we can say that it would compensate the poor electronic connectivity but yes, the sodium does it. And the fact that you have very small electron transport is the reason why you will have so well defined redox couples positions. You know, really titanium four plus three plus is always at the same place for given phosphate. So if, and if you mix titanium and vanadium you will have both couples. They will not mix somehow. You know, the electrons are really step by step into the different and this is because you don't have electron transport between the octahedral. Yeah, it's a very interesting system. The next question is on the two-phase behavior in the plateau you observed additional phase transition that did not manifest in an extra plateau. So this question is for me actually. I think it's very interesting. That means the energy difference between the phase must be very small such that the voltage is almost the same. Is that your interpretation as well? Yeah, yeah, yeah. And it has been definitely and it has been even modeled somehow by DFT by Pierre Emmanuel Canepa with whom we have been working hard on that. So this is really such a small difference, you know. So we're trying to find some ways of seeing it electrochemically. There should be some signatures somewhere somehow but with classical gavano static data like that you see just a very boring flat plateau. But yeah. What is fascinating is this kind of behavior same for NVPF happens in materials that show very, very high rate capabilities. No, you have all those many different and I think it has been mentioned in very old papers by Atsuo Yamada, you know that when you have this sort of transient phases this really favors fast electrochemical operation between the cathodes, right? And somehow maybe the existence of this intermediate phase is maybe one of the reasons why we can go so fast from any three to any one through this intermediate one. It's same point, same point in VPF actually. This reminds me of some of the models on LFP as well. Yeah, exactly. This is what I meant. This is when I mentioned about Atsuo Yamada it was because what he did on LFP, yeah, some time ago, which you would have, of course. I forget the name. Is it Domino Cascade? I forgot the name that was coined for this interfacial transports between phases. Yeah, the Domino Cascade model was presented and invented somehow by Krudelmas and Lohans Krugeneck. But it's not exactly what Yamada was talking about. Yes. Great question. So I think now because we only have a little bit of time left, I suggest we now have a discussion also with Jerry. So Jerry, if I can ask you to come back. So typically in our seminar at the end, we try to coalesce the perspectives a bit. And I think today we are very honored to have both the academic and the industrial perspective and everything in between. So I thought we can devote the next 15 or minute or so to talk about intersection between the two. And I thought maybe I'll just start with a question from my side, which is, Christian, in the many of the materials that you have presented and also in yours Jerry, if you were to draw a line between the fully charged and fully discharged states and then calculate the average slope for many of the sodium compound, it's quite a bit larger in terms of the slope, meaning that as the battery operates, it may have the same energy density, but the range of the voltage is greater. And I wanted to ask maybe Jerry for you to comment on product quality here. So as you increase the average tilt of the voltage, whether it's through multiple plateaus or other things, does that present any commercialization challenges when it comes to integration in the system? And also has the electronics change over the past, say 10 years to better allow, say, car manufacturers to work with this wider voltage range? That's a good question. Well, the voltage slope in a layered oxide hard carbon, it's a double hit, right? When you compare it with, say, LFP and graphite, which are both very flat voltage profiles. If you look at the hard carbon profile, right, it's composed generally of kind of two aspects. There's a slopey part and there's a flat part. And then on top of that, the layered oxide is generally sloping. It depends a bit on what transition metals you use clearly. But if you go for something like we've got, which is operating on the nickel two plus four plus couple, it's generally more sloping than an NMC or an NCA. So we end up with our voltage range is from fully charged in the full cell, 4.2, fully discharged is somewhere between 1.51 volt. It's a long voltage range. We prefer not to have that well, right? Cause that kind of hits a few things, but it makes it more difficult in terms of the electronics and the BMS to be honest. But again, if you look at it, you don't lose, let's say in an application, you cut off at 2.5 volts, right? Cell voltage on discharge. There's not a huge amount of energy below 2.5 volts because the very aspect is it's at a low voltage. So we overcome a lot of it by just cutting off at something like 2.5 volts. We would prefer though for a lot of applications to have it less sloping. I mean, there's a long answer to your question, but... Yeah, I would mention indeed your question, William, that this is true for the leotoxides and the thing it's true for both lithium and sodium and leotoxide. For the polyanions, you have basically most of the time much better defined plateaus, I would say. Let's say the case for instance, for NAVPO4F, very flat plateaus at very high voltage and you don't have these voltage ranges. So it's not... I don't think it's due to the sodium chemistry, the structure and the structure is layered versus polyanions in this case, to answer. You know though, Christian, a lot of it, it's the combination of the cathode performance and the anode performance, right? So, you know, I mentioned in my talk, well, about there's a lot of work. You know, when we first started commercializing, we basically just took commercial hard carbons, right? And commercial hard carbons were developed for the lithium ion industry, not for the sodium ion industry. And what those had, the way they were produced meant that more than 50% of the discharge capacity was in the sloping region, right? Which is not what you would choose for an energy application. But now we know what those mechanisms of insertion and extraction are in the hard carbon, such that if you make the hard carbon in a particular way, you can have pretty about all plateau, right? So you take away the sloping part and that just increases the two things, both the average operating discharge voltage of the full cell and the less slopey discharge profile. So you can overcome some of these by chemical engineering of the hard carbon. Christian, just to come back to your point about the slopiness and what is the cause of it, my feeling is at least in the NASA constructors, usually you have to employ more than one electron redox. And I think intrinsically, you will have to deal with the two voltages there. But I think you're right. There should be no specific dependence on lithium versus sodium. I think it's just the use of the NASA constructor generally requires more than one electron redox. And that is something that you have. In this case, you have to play with more than 1.5 volt difference from the V3, V4 to the V3, V2, for instance, couples, yes. So coming from the chemistry side, Christian, I mean, this is a very challenging task of using multi-electron redox, but not wanting them to be too different. Do you see also tricks on the chemistry side that we can play to bring the redox potential a bit closer together? I've been thinking about this for a long time and couldn't find a good answer to it, but I thought I would poke your brain. What happens in the NVPF? And I don't know if we can do that for the NASA cons, but what happens in the NVPF is kind of fascinating. You have lithium-3 plus in this by-long-term unit, but then you end up with VO, the vanadyl O2 plus, which is very different than V3 plus, I mean, you would say, and that ends up to slightly different voltage. And if you have these two units and VO2 plus can be seen somehow as V4 plus, for the NVPF O2, basically you start with, you would say with vanadium 4 plus, but you have basically the same operating voltage than the NVPF, which is vanadium 3 plus. So there's a lot to do still here on maybe answers a little bit of your question, but we haven't really seen that in the NASA con yet because the structure is different and the creation of this very short vanadyl VO bond really is detrimental for the NASA con cyclability, which is not for the NVPF. Well, Jerry, maybe the biphysia could be an approach to work this and it cannot be done in a single crystal structure, maybe can then done in two to bring things a bit more closely together. Possible, yeah, possible. You know, we've got the same issues of the native and the NAVPO4, I think that we're going nickel 2 plus to nickel 4 plus, right? When we charge and then 4 plus, 2 plus and discharge. So it's a multi-electron transfer on the nickel, which is good, right? It's very efficient in terms of how much nickel you use in the cathode material, but you just end up with this very wide voltage range. But it's a tough one to overcome because there's not that many other transition metals you can use. We don't really want to go down the route using some of them like cobalt, for instance, or chromium. So you tend to keep on coming back to a combination of nickel, iron, manganese, and maybe copper. These are the ones you generally use in layered oxide chemistry. And the biphasic gives you another degree of freedom because P2 and O3, for instance, like we use up for the same redox couple, operate at slightly different voltages, but it's not enough to smooth out the profile completely. Jerry, on the biphasic approach, and I know that blended has been explored very extensively industry over the past 10 years. Are you concerned, as they do in blended electro-differential degradation between the two phases? Everything looks good initially, but how does things look later on? Yeah, yeah, definitely, absolutely. You've got to be aware of that all the time. Yeah, I would say that, like I said before, though, that the synergistic effect we see is that physical mixtures are not the same as chemical mixtures because of the integrals. It's a marked difference. It's maybe like a factor of two or three in terms of the cycle line. It's a big difference. We naively thought when we started this work that, hey, if you, we know that P2 materials are generally good for power applications, right? Because of the prismatic position of the sodium and the O3 had higher specific energies, but we're better for energy applications, not so good for power. So we thought, well, in an ideal situation, we could have a P2 silo and an O3 silo. And if someone came along and said, like, I want this particular application with this power and energy requirement, you could just mix those two and well, you could do that, but it's better to make it chemically. That's what we found out because of the integrals and the synergy between the two. But the beauty of it is that as I said to you earlier, you can take a precursor mix, treat it differently and end up with mixed phases of O3 and P2 with markedly different amounts of each with no impurity phases. It's very powerful. So that gives you, in theory, give you a power cathode material and an energy cathode material from the same precursor mix. It's actually pretty good. I agree. Well, we have a couple of minutes left. Christian and Jerry, both of you are involved in commercialization of sodium on batteries, Christian, and your work with Tia Matt and of course, Jerry, with Faridion. As I mentioned after Jerry's talk, I think in terms of the academic side, say research side, the results are extremely encouraging. So I wonder if we can close by talking a bit about what are the remaining gaps needed? Either it's fundamental or more on the development side or scale of side, that is really needed, a sort of an honest look at the remaining challenges to make this happen. And maybe I can ask Christian to start. If we're talking about sodium ion batteries versus lithium ion batteries, one of the very important thing is the foreseen cost, right? And the cost of the cost per kilowatt delivered, whatever. And so the challenge is to identify lower cost for me, lower cost cathodes, I would say. I think Jerry mentioned that as well. That means trying to move towards iron manganese materials and stabilize them. I think this will be a must for to make sure that sodium cathodes will be of very high interests for massive batteries, for instance, for this kind of things. Yeah, you need cheaper batteries because as Jerry mentioned that one of the main competitor to sodium ion technology was LFP, right? Somehow the sodium ion technology is comparing their own performances versus LFP, which is hard to beat in some cases, but which in terms of energy density could be beaten. And one of the challenges that Jerry maybe didn't mention too much in his talk is about the, for the layered oxides, the stability in a humid atmosphere or air about the layered cathodes. Maybe this is also a big challenge. Myself, I'm not able to address it because I don't work on the other sides, but maybe Jerry can comment on that. Yeah, yeah, no, it's a good point. Again, the ambient sensitivity depends on what transition metal combination gives Christian as you know. And there are things that you can do. If you look at an as made nickel based O3 layered oxide, then the open circuit voltage of the as made material versus sodium comes out to be about 2.6 volts versus the sodium reference. If you know anything about it, you'll know that's very marginal moisture sensitivity, ambient sensitivity, but you can shift that with the right combination of substitutions. Maybe I'll leave it at that. You probably know, I'm sure you know that, Christian, right? There's certain things you can do. But go back to your original point there, well, about what we need to do moving forward is I always say to the guys at Friday, right? When you start to try and commercialize a new battery chemistry, right? There always used to be the three things you had to check, the three boxes you had to check, right? You always knew I had to check boxes of performance, right? Cost and safety, right? All those three things you had to check. There was no compromise, right? But now you've got to add in sustainability. If you're trying to license or develop a new chemistry, you have to check that box as well. So I think moving forward, well, I think I mentioned it in my talk, I think if sodium ion batteries could move to a zero nickel layered oxide cathode, we would now have squared the circle, right? We would have a completely sustainable high performance cell chemistry. It'd be beautiful, right? It was based on iron and manganese. You've got it's low cost, it's high performing, it's sustainable, it's kind of got everything. That's easy to say and difficult to achieve, but it's only that one delta that needs to be done, right? I say only, it's not easy, but it's got better opportunities, I think, than pretty well in the other cell chemistry. Yeah, I completely agree. And I think, you know, it's very, we're very fortunate to have you both here because you came from the very original days of LFP. We look at the drivers for LFP's resurgence is really safety and cost, right? And that delivered. And the technology hasn't changed very much beyond those. You know, well, I said this to Christian before, right? When Valence first introduced its first LFP product, right? This is 2002, right? We were literally doing cartwheels and we got to 100 watt-hour per kilogram. That was our target, right? Now you see cells that are like 200. It's the same chemistry. It's the same graphite LFP. And it's just moved on because people have done a great job in engineering the cell. They've done a fantastic job. And when you see it now, when you're dealing with the new, when you're doing silicon graphite composite anodes, they're talking about 225 watt-hours per kilogram. I never would have believed it well. I would have never have believed that was possible. 225 is crazy, right? We were happy at 100 when we first did the work at Valence. So great engineering a little bit at a time and you had them. Fantastic engineer. You have to say, right? We did the chemistry maybe and they did great job on the engineering in a large format. Great, great job, right? No. Well, you know, maybe I can paint my own dream. Given today's lithium carbonate, lithium hydroxide pricing, I think if you can simply replace the lithium in lithium-iron phosphate with sodium, then that's a done deal, right? Everybody would do it. But that's, it's not so simple. It just doesn't work for you. Christian knows it's better than anyone, right? It goes against you. Nature goes against you when you just try and make that by solid state. You know, you have to do it. You have to both do it by a different route, you know? So, again, maybe there's still room in crystal and I'm still optimistic in finding new materials and new properties, et cetera. You know, this NA2, OK? And then I showed you that we have a new compound where that operates at higher without any extra elements than N, NAV and P. So you can play with many things to try to change the electrochemical properties. There's still room for the crystal chemist, that's fine. And then the engineers will make it also work better. Absolutely, absolutely. And we will not end today's talk by asking everyone to comment on the timeline. The technology will come soon at some point. But I think the optimism is really high, at least in this seminar. Christian and Jerry, I want to thank you both for taking the time. I know it's late over there. And I hope to see you both soon in person somewhere. Sounds good. Thanks, Will. Thank you very much. And Justin, can I have the closing slides? And we do have one more seminar before the winter holidays. Our final seminar, which is two weeks from today, will feature Deborah Rawlison from the Naval Research Lab in Veronica Augustine from North Carolina State. And they will talk about aqueous energy storage of a grid storage applications. And again, thanks from Stanford for everyone joining today. I hope you learned about sodium ion as much as I have. And please stay connected with us. Thank you very much.