 Good morning from a smoking bay area in California. My name is Will Chu. I'm the co-director of Storage X Initiative at Stanford University. And along with my co-director, Professor Itui, it's our pleasure to welcome you to the Storage X symposium. So we're doing something new today and for the next several sessions, we're calling this new set of talks, Storage X X Equals Question Mark. As many of you can appreciate, there are a variety of ways to store energy. And the next few sessions will precisely cover all the other options of storing electricity beyond battery technologies. And to kick us off today, we're going to have X Equals Fuel. Tom Jaramillo and Susanna Haley will be given talks on fuel cell and electrolyzer technologies to convert electricity to fuel and back. So let me briefly introduce our speakers. Tom is a professor of chemical engineering here at Stanford. And he's the director of the Suncat Center for Interface Science Catalysis. And that's a joint center between Stanford University and Slack National Accelerator Laboratory. Tom is a pioneer on the design of electric catalyst, which is guided by theory and tightly integrated with experiments. Tom is also a senior fellow at the Precourt Institute for Energy here at Stanford. Susanna Haley is a professor of material science and engineering at North Western University, where she's also the faculty co-director of Northwestern Institute for Energy and Sustainability. Susanna spent 18 years at Caltech, where she was also my mentor before moving to Northwestern four years ago. And Susanna is a pioneer in a variety of fuel cell technologies. She's the inventor of the solid acid fuel cell and has been working also on high-temperature solid oxide fuel cells. So between Tom and Susanna, we're going to get a very broad coverage of fuel-based technologies for energy storage. And I invite you to listen in and learn more about X equals question mark starting with fuel. So Tom, if I can have you on the stage, we're excited to hear your talk. Tom. Thank you very much, Will, for the kind invitation to be here for the kind introduction. It's really exciting for me to be able to participate in this symposium, which has been a great success. So again, congratulations to you and E and all those who are organizing these symposia, really to educate ourselves on storage options. And as you said so eloquently indeed, today we're going to be shifting gears a little bit from previous symposia topics. But still within the very important themes of electricity, renewable electricity, and storage. And so that's definitely the foundation for everything I'll be talking about today. The perhaps differentiating factors we're going to be talking about using that electricity as a storage means in the form of molecules of molecular bonds. So particularly we're going to be talking about storing it in hydrogen, storing it in carbon-based products, and ammonia. So without further ado, maybe I can get into maybe one of the most important points to make, and that is that everything I'll be talking about today is a collaborative effort. It's really tremendous to be part of the Suncat organization, the Suncat Center for Interface Science and Catalysis. As Will also pointed out, it is a partnership between Stanford and Slack, where we have a great number of researchers who are applying their trade towards these very important missions. And specifically we want to understand interfacial phenomena. We want to develop a science-based design of solid catalysts and really provide solutions for sustainable processes for fuels and chemicals. And I'll explain what we envision in that space. We are a theory-centric center where things are grounded very much on theory, where our computational colleagues are really doing calculations on how reactions proceed on surfaces for a variety of different chemistries that really informs a lot of the synthetic chemistry that we employ within the center to synthesize all kinds of different materials with different compositions, really using the periodic table as our playground, making different morphologies that could be nanowires, mesoporous materials, nanoparticles, etc. And then we go about characterizing them not only with the conventional tools of one mind fight at a university such as Stanford, but also at the wonderful, powerful tools that we find at Slack, namely a synchrotron as well as a free electron laser. Then we can of course put these materials and catalysts to the test, really evaluate what works, what doesn't, feed that back into the theory effort and continue to iterate as such. Data science is also of course a very prominent theme in this space and we are trying to capture all of our computational and experimental data in a way that has previously not been done before to ask ourselves a question, can we learn new things and complement the physics-based modeling that we typically do? So this center is funded by the U.S. Department of Energy Office of Science and really built on the bedrock of wonderful administrative staff. I want to point out as well, I meet that Bujan Harbin who allow us to do what we enjoy every single day. Now, let's talk about fuels and chemicals. So the modern fuels and chemicals industry in my estimation is one of the greatest achievements of humankind. It really is something marvelous and spectacular. If you take a look at just some of the molecules that we produce across the globe, and let me tell you that we produce about 70,000 different molecular products for commercial use across the globe, naturally different types of molecules, different functionalities, it could be antibiotics, it could be fuels, it could be plastics, there are so many different possibilities. And this is just an example of four of them. These are some of the bigger ticket ones. Take hydrogen, for instance. Hydrogen is produced at a rate of 65 billion kilograms a year. And if you just divide by the seven or so billion people on earth, that's about nine kilos per person, really an absolutely tremendous scale of production. And where does this come from? It comes from fossil fuels. It comes primarily from steam methane reforming, which is a large CO2 emitting process, as you can see here. And people talk about hydrogen potentially being a fuel for the future and maybe an energy carrier. All of these open questions, well, when you look at numbers like this, it's pretty clear that hydrogen plays a very important role in our global economy already. Where does that hydrogen go? It goes into other very important processes. We don't typically go to the store and buy hydrogen off the shelf, but we do buy products where that hydrogen ends up into those products. For instance, it goes into a fertilizer and specifically ammonia. So ammonia is another big ticket molecule. We produce it 180 billion kilograms a year. This is the Haber-Bosch process. There are a few hundred Haber-Bosch facilities across the globe that take the nitrogen from the air that we're all breathing right now. And that takes the hydrogen that comes from typically steam methane reforming on the left, and it reacts it over this iron nanoparticle to make ammonia. Once you have the ammonia, you can make fertilizer in different forms. You put it into the ground, you grow a crop, and that's how we get the fixed nitrogen into our body. You either eat the crop directly or you eat the animal perhaps that eats the crop. Either way, half the fixed nitrogen in your body, half of that fixed nitrogen touched one of these iron nanoparticles sitting inside one of these facilities across the globe. Another spectacular process that we can make it at the scales needed to feed billions of people across the globe and do it in a cost-effective manner. Other products here we're looking at, gasoline, plastics. I actually used to work out here in the Houston Ship Channel making monomers at the megaton scale. Plastics, clearly a lot of room for reinvention. We're already making 300 billion kilograms a year of plastics and climbing and lots of opportunity to reinvent those processes. And then gasoline on the left, if you look at this scale, one trillion kilograms a year. Now we're talking another order of magnitude in scale, really. This is an ExxonMobil refinery in Baton Rouge. It's just one of 700 or so oil refineries across the globe, and each one is cranking out about 100,000 barrels of oil per day. So if you just take a look at the scales of these processes, it's really tremendous what the scientists and engineers have been able to accomplish over the past several decades in fact, centuries to be able to not only make these molecules that are extremely important in our world, but also to do it at scale and at a cost-effective rate. Now there are some challenges as wonderful as these processes are. They're not perfect. They are, for one, not sustainable. They're all coming from fossil fuels and almost every process on earth ultimately comes from fossil fuel feedstocks, a few don't, but most do. Another issue is that while they do access billions of people on earth, not all billions of people on earth have the same access to these types of molecules that we might have say in North America or Europe and other parts of the world. So there's an opportunity to really come up with new processes that have that ability to reach the other people. So we move forward. What's the opportunity to change things? This is really the game-changing scenario. This is something I think we're all familiar with, and that is the dropping price of renewable electricity. No need to belabor the point here. I think we're all familiar with the fact that renewable electricity is here. It's growing. It's market penetration continues forward, and all the meanwhile, just in the last decade or so, we see tremendous drops in price for both wind and solar, reaching some very interesting values of say four cents a kilowatt hour on average and with the possibility to continue dropping ahead. And the question is, how can we utilize that low-cost renewable electricity to perhaps power renewable sustainable processes for the production of fuels and chemicals? Can we make those same types of molecules we saw in the previous slide, but instead of powering them with heat and pressure that's coming from burning fossil fuels, can we power them with electricity instead? I've got good news for you. We already do that today, in fact. So if you take a look at some processes that are already scaled up, this is not new technology by any stretch. You can see a photo here from plants in Louisiana in the 1950s. This is an aluminum electrorefining facility. I remind you that bauxite is the orb by which we grab aluminum. It's in an oxidized form, and so it's really electrochemistry that reduces that aluminum ultimately to get that aluminum oxide to aluminum metal. And you can see the numbers here. It's also at the 60 billion kilogram a year, very large scale. It comes out to about 100 gigawatts of electricity continuously going into this process to be able to meet the world demand for aluminum. Another example is chloralkylide process, otherwise known as brine electrolysis. So it's basically a very highly concentrated sodium chloride solution in water, and you're using electricity to do the chemistry to make sodium hydroxide and chlorine. Again, that 60 billion kilograms a year type of scale, 11 gigawatts going into this particular process. And then there's water electrolysis. It's also old school technology. It's been around for a really long time, commercialized for about a century. Now we're looking at a plant in Norway from the 1950s. Back in the day, this is how they made their hydrogen that used to go into the Habervasch facilities with steam methane reforming and the dropping price of fossil resources. Then that became a cheaper way and still is today a cheaper way to make hydrogen. The question is, maybe we want to come back to this mode given renewable electricity prices, the way that they're going and the other driving forces for improved sustainability of processes. So this is coming back into fashion. Now alkaline electrolyzers are really the the in-ground tech that is deployed around the globe, but PEM electrolysis is advancing very fast and many believe that that's going to have a brighter future ahead than continuing to build off of the conventional alkaline technology. So this is really to show that the electrification of chemical manufacturing is not just pie in the sky. We literally do this every single day to make very important molecules like chlorine, sodium hydroxide and aluminum. So how do we create a new paradigm? This is some top-down techno economics basically saying the production cost of anything you might want to make any molecule in dollars per kilogram is going to be a function of the capital expenditures, of course, and the operational expenditures. So these are four targets that I just want to bring to your attention, highly ambitious. But if we can accomplish these four things, then we can flip the entire chemical and fuel manufacturing industry in a different direction. We can revolutionize it. And so having renewable electricity prices is about one cent a kilowatt hour, I think, is a game-changing value. At that point, the energy input that would go into a chemical process, the electricity cost are highly manageable. I remind you that production costs of a lot of these molecules that we've been talking about are kind of 50 cents to a dollar per kilogram is a typical price. And if you think about what can you go to the store and buy for one kilo, you pay only 50 cents to $1. There's not much out there. And so these are really, really low-cost targets to hit. And so you need inexpensive electricity to go into it. Chemical processes like to run 24-7, 365, and renewable electricity does not come in 24-7, 365. And so this is where energy storage can really play an important role. Even time shifting on the order of just hours, half a day, a day could actually make quite a bit of difference. But naturally, we'd like to go with longer-scale grid storage options on the order of more months or seasons. And so if we could hit, say, $10 a kilowatt hour, then we could totally time shift that electricity and allow these processes to run continuously. Now, of course, that increases the cost of the overall operation. And that's part of the trade-offs that we need to contemplate. But naturally, this is a very ambitious target. Carbon capture is also important. If you want to make a carbon-based molecule, and you do not want to get it from fossil fuels, then getting it from CO2 would be a really nice place to get it. Another ambitious target here, there are many different sources of CO2. Getting it from the atmosphere would be the best. But that's also the hardest. And so there are, thankfully, a number of point sources of CO2, bioethanol refineries, for instance, cement manufacturing, natural gas reservoirs, where it's possible to get CO2, perhaps has a better chance to reach these types of low costs. And then finally, you need the box that you feed all of this into. You need to feed in electricity. You need to feed in your feedstocks, water, CO2, nitrogen, what have you. And the cost of that technology itself that's actually doing the chemical transformations for you needs to be inexpensive. And if you take all these numbers together, that's how you're going to get to say production costs of around 50 cents to a dollar per kilogram, which again would then be directly cost competitive with conventional processes that we use today. So these lots of work for all of us in this domain, there's no one single lever to work on from an R&D perspective that's going to get us to the finish line. We have to work on many different areas of technology, integrate them together. The things I'll be talking about are really centered in this box here on the bottom right. So the key is that it's going this is how it all fits or at least some schematic as to how things might fit. At the end of the day, we want to take renewable electricity or renewable resources to power electrochemical processes to do chemical transformations. You can either run electricity directly, you can use the solar resource to drive things with photon driven chemistries. And in some cases, you might be able to make the product you want directly, like you might be able to feed in nitrogen and water and be able to make ammonia fertilizer right off the bat. Or you can make hydrogen, for instance, and feed it into more conventional processes for Haber-Bosch that then makes the ammonia. So the bottom line is what you need, fuels for transportation, chemicals, materials, fertilizers, what have you. We just need to rethink not only what new areas of technology we can build, but also how they couple together and integrate with one another and integrate with conventional processes that exist today. Now in this period, there are three examples I want to bring to your attention on three types of molecules that where this could perhaps play an interesting role. And again, that's hydrogen carbon-based products and ammonia. We'll start with hydrogen first. So here's some more technical economics. This is an analysis conducted by Proton Onsite. It's a company in Connecticut that specializes in PEM electrolysis. In fact, they were recently acquired by NEL. NEL is formerly North Kedro. I showed you a photo of one of their banks of water electrolyzers from the 50s. And in this concept plant, making 50 tons a day of hydrogen, they costed this out. They did the chemical engineering 101 plant design, which says you start with the reactor, which is your electrolyzer. And then you need other components, balance of plant systems, you need power supplies, you need water management, you need hydrogen management. So they costed this out, and they basically said, you know, they saw a trajectory towards getting down towards a 50 to 60 cents per kilogram of hydrogen, cap-axe, which is really exciting because if you give a facility like this free electricity, 24.7365, this would lead to cheaper hydrogen than the, say, $1.20 that we typically pay in North America per kilogram. Unfortunately, electricity does not come in 24.7365, and it's not free, renewable electricity, I should say. And so this is why on my previous kind of top-down technical economics, I was suggesting more of a 20 cents per kilogram is a better target to hit. So how are we going to continue to lower the price of these electrolyzers? You can see in the pie chart here on the left, this is where the dominant costs are coming from. More than half of their costs are coming from the stacks themselves, the electrolyzers, and a big chunk of that is coming from the precious metal catalysts that are needed in pan-electrolyzers, namely platinum and iridium. So this has motivated a lot of our research groups where it can try to reduce precious metal content, both on the anode and the cathode, to either completely eliminate or at least reduce the platinum and the iridium. So if you want to make a catalyst, and now we're going to do a deeper dive into the tech space here, what are we doing scientifically to address these concerns we can make for lower cost systems? You want to make a good catalyst and say you don't want to use platinum. A good place to look is biology. Biology is, of course, highly evolved. We have these enzymes all over. You look outside, you look at a tree, you look at a plant. They're working inside of our guts right now, maybe chewing on the breakfast that we had this morning. Bottom line is, biology has come up with a really good system to do chemical transformations. And here's hydrogenase and nitrogenase, which are very good nature enzymes, natural enzymes, for producing hydrogen. You can see here now they're every bit as good as platinum, but there's no precious metals in sight. And so thanks to calculations done by Barrett Hineman and Jens Norsko some years ago, they really try to understand what makes these things tick. And the long story short is that the motif you see here, which is an undercoordinated sulfur, behaves very differently. And it's far more active a catalyst in the enzyme than say the sulfurs that are up here that are coordinated to three metal atoms. So back when I was a postdoc at the Technical University of Denmark, working collaboratively with Jens Norsko with Barrett, with a number of experimentalists, including my advisor, Eve Corkendorf, we were really inspired to look at this material. This is molybdenum sulfide. You might have seen molybdenum sulfide in various forms. It's a layered material, has a lot of interesting material properties. It's kind of like graphite where the Van der Waals forces that bind these sandwiches is a sulfur molybdenum sulfur sandwich that is stacked to other sandwiches via Van der Waals forces provides for good lubrication. It's also a very important catalyst using oil refiners around the world. We can thank molybdenum sulfide for hydrotreating processes that get rid of nitrogen and sulfur out of petrochemical or petroleum feedstocks. And that's why we can inhibit or mitigate acid rain, is thanks to this catalyst. Now this motif at the edge has sulfur, as you can see, decorated in a way that is much more like this sulfur here that is active in the enzyme compared to say the sulfurs on the top and the bottom. And so we went about synthesizing nanomaterials to see if in fact the edges were the active site. And they were. All the activities turned out to be very active catalysts. All the activity happened at the edges did not happen on the basal plane. And with that is a scientific foundation coming to Stanford to start a research group. Some of our early stage researchers, Yakov Trebo, went about synthesizing all kinds of different nanomorphologies of MOS2, really trying to get as many edge sites as possible to expose. And so the long and the short of it is the more edge sites you put into your electrode based on the different formulations of molybdenum sulfide, the higher activity you would get. And so here's a polarization curve to show you some data. This is in sulfuric acid. The equilibrium potential for the reaction is of course at zero volts versus the reversible hydrogen electrode. And what you want to do is as much current as you can possibly get as close to the equilibrium potential as you can. You can see from formulations one to two to three going from nanowires to mesoporous materials to these small molecular clusters. We're able to march along and operate closer and closer to equilibrium. Now these are very fundamental studies that showed that there was a possibility to engineer the catalyst and the electrode for improvement, but 10 milliamps per square centimeter is not going to cut it for an electrolyzer technology. And so here's where my PhD student Desmond took a different approach basing his work off of all the fundamental knowledge that we were working on. He said, can I reformulate these into a more device-friendly concept? So we built our own home-based electrolyzers. It's a very modest MEA you can see here. It's only five square centimeters where Desmond took a lot of these sulfides and phosphides and fossil sulfides, different transition metals, many that are not depicted here, really put them into ink formulations to start making devices and see how they would perform. So here's the current voltage curve. Now we're drawing one amp per square centimeter, which is a current density that is relevant for the commercial grade tech. And not surprisingly, the platinum iridium is the best. We know that platinum and iridium are the best catalysts for use of these types of devices, so that's not surprising. But what was really good to see is that these different non-precious metal formulations, when you replace the platinum with them, they still work. They still can get you your amp per square centimeter. And lo and behold, you have to pay about a 200 millivolt penalty to get the same current density. This is not surprising either because no human on earth has ever made a catalyst that's as good as platinum without containing precious metals. So that does continue to be a mission of ours to try to ultimately make non-precious metal systems that are as good as platinum. But even though we're not there yet, this still is a trade-off worth considering. Basically ask yourself in a market where the cost of electricity is very inexpensive, maybe it's worth your while to have a lower efficiency electrolyzer and you pay a little bit more electricity, but it's a free catalyst compared to platinum. So it still might be interesting in a variety of markets, but naturally want to continue trying to get these curves marching towards platinum and not have to pay even a millivolt extra if possible. Now we want to take this one step further and that is to see how things would perform in a real commercial water electrolyzer. So that company I mentioned earlier, Proton Onsite, again recently acquired by Nell, we've got a project funded with them through an SBIR that was to really test these types of catalysts in their commercial grade platforms. So McKenzie and Lorie went about synthesizing and scaling up some of our catalyst synthesis and shipping it off to Connecticut where they would integrate it into their electrolyzers. They do sell this electrolyzer. It's a commercial device, but of course it has platinum and iridium in there. And so they basically just kind of swapped out one catalyst for the other and looked at the performance. And now we're operating at 1.8 amps per square centimeter at 400 psi. 50 degrees C is exactly the operating conditions of their commercial system. And lo and behold, the polarization curve show that yes, when you have a non-precious metal catalyst, in this case it's a cobalt phosphide based system, you do pay again the 200 millivolt penalty. That's exactly what we would expect it to do for the reasons I mentioned earlier. That nevertheless, it was able to hit the performance metrics that they were after. And then the question was what about long-term durability? And so they just ran this thing for about 1800 hours before it was time to shut down the experiment because the project ended and we saw absolutely no signs of degradation. It was just absolutely a very smooth operation all the way through. No catastrophic failure at the end. I am curious how long it would have lasted, but that wasn't in scope for that particular project. And also handled some intermittent operation, some unintentional power shutdowns. And yet the catalyst was able to continue moving forward, which is a nice, well that was not intentional. Boy, that was nice information to see that really lends itself well to commercial technologies perhaps plugging into variable sources of electricity. So now with hydrogen, so this was an example of really showing, kind of going from fundamentals of catalyst development out to more applied commercial grade type tech. This hydrogen I do see is really the flagship of all this electricity to X where X equals fuels and chemicals and is leading the path. We want to do the same with other chemicals out there. And carbon-based chemical is obviously massive in our world. We want to be able to work in that space as well and develop similar type of electrolyzers, high performance that can convert CO2, let's say to any desirable carbon-based product you might want with CH and O. And there's other molecules too with ammonia, for instance, being another example that we talked about previously. So let me shift gears and let's talk about carbon and then we'll talk about ammonia and then we'll wrap it up. So one of the big differences in CO2 electro catalysis versus hydrogen production is selectivity. So if you throw CO2 in water at an electrode and you want to pump electrons and protons onto that CO2, the challenge is that you can go a lot of different places. This is just some of the molecules you can imagine making. And steering that selectivity is a massive, massive challenge because any commercial tech where you're building a CO2 electrolyzer, if you have to do separations processes after the electrolyzer, it's going to be really tough to compete cost-effectively. So you really want a catalyst and a device that can give you what you want, the products you want, selectively, high purity, and not make the products that you don't want. And oh, by the way, in this case, we don't want hydrogen. That's a major problem because the hydrogen kinetics are generally a much easier, more facile than CO2 reduction kinetics. So we also want to be able to inhibit that. So a lot of our work has been steered in that direction. And there's a lot to say. And I will try to keep this story short and sweet. That is, first and foremost, it's been wonderful to work with this tremendous team of scholars working in the space over the years. I want to give shout-outs to our initial waves of researchers, Kendra, Natasha, David, Toru, Jeremy, Steph. These are really establishing the foundations, these PhD students in this area. And Natasha and Kendra, in fact, went off and started a company called Opus 12 built on CO2 electrolysis tech. I'll also be pointing out some work from our postdoc, Le Wang. And also our former postdoc and now staff scientist, it's like Chris Hahn, who is central to all of our CO2 electro-catalysis studies. So some techno-economics to get the ball rolling here is you look at the difference in CO2 electrolysis and hydrogen is, again, you're making different types of products. So the techno-economics for every type of product, be it carbon monoxide, be it ethylene, be it ethanol, be it propanol, whatever it may be, the techno-economics might look different based on the market. And this is just a way of saying, when you look at a plot like this, there are many, many assumptions in these models and many factors to consider. But one thing that we know we need is low electricity costs, as I pointed out before, and high energy conversion efficiency. And the bottom line is that no matter what molecule you look at, there is a profitable region in which you can operate if you can develop the tech in the lab. So the techno-economics work out. Basically, the answer is yes, it is possible for electrochemical technologies to impact the fuels and chemicals industry if we can develop the technology with the performance that we're hoping for. So with that, then let's take a look at some of the catalysis here. So some of that initial work from Kendra, Tasha, David, you can see that they were feeding CO2 here. It's a simple copper catalyst at standard temperature and pressure, 0.1 molar potassium bicarbonate, and amazingly, they detected 16 different products of reaction. We had really engineered our devices, our reactors to be sensitive to identifying, quantify the products of reaction. The good news is that copper can make all these really interesting products, but there's some challenges. First of all, it's a big mix, and I mentioned separations is a challenge. So it's totally not a selective catalyst. That's a problem from a commercial technology. Another issue is that you have to apply very negative voltages. The equilibrium potentials for most of these reactions are around zero volts versus RHE, so similar to hydrogen, and yet you have to dial up the voltage to about negative one volt before you get these chemistries to go at a reasonable rate. Another issue is that a lot of the more valuable products here that are easier to mark at entry points are oxygenated molecules, things like, say, an ethylene glycol or an acetaldehyde, but oftentimes, the hydrocarbons are what are favored, like ethylene and methane. Now, those are great molecules. I cook with methane every single day on my stove top, but those are also really inexpensive. The hydrocarbons are the cheapest of the bunch, and so maybe the most difficult market entry point. So really just to say that there's a lot of challenges in steering the selectivity, but at least there are catalysts that are capable of doing this reaction. Now, we've been thinking a lot about mechanistic possibilities here of how to steer the chemistry, and we're not, we're certainly not going to have the time to go through all of this in great detail. If your eyes are hurting a bit by staring at so many different possible pathways and so many possible intermediates, let me say that this is just a distilled version of the possibilities. There are many more possibilities that are not depicted, and the complexity is really what I'm trying to get across. The more electrons and protons you pump onto CO2, the more challenging it gets to understand the mechanism. So we've been trying to answer lots of important questions like, how do you suppress hydrogen when you're trying to do CO2 reduction? How do you favor C-C coupled products? How do you favor oxygenates over hydrocarbons? How do you potentially make a single desired C-C coupled oxygenate? So we've been focusing a lot on these efforts, and I just want to show you one vignette, and then we'll say a few words about ammonia and wrap it up. And that is that one thing that's really interesting about CO2 electrolysis that we came to learn is that electrode surface area can impact selectivity, which is interesting. When you think about electrode surface area, you think of the higher electrode surface area, the higher the reaction rate, which is natural. That's kind of what these blue lines depict, as you dial the voltage more negative, you should get a higher current density with increasing roughness factor of electrode that we all are aware of. But what we also came to learn is that the further away you are from equilibrium, the more you're going to favor the reduced product, meaning the hydrocarbons. So if you go higher surface area electrodes, you should be able to operate closer to equilibrium and thus be able to get a higher ferritic efficiency for oxygenates. This is the hypothesis that drove a lot of the work that Leigh, Stephanie, and Chris had been working on. So we went about synthesizing just really high surface area copper that frankly we borrowed from the supercapacitor literature. These are nanoflower type morphologies, as we like to call them, and without belaboring the point, very different than what we saw with the CO2 electrolysis on a plain copper foil some of our original work. Now we can operate, this is now CO reduction as opposed to CO2 reduction, we can operate very close to equilibrium, less than a quarter of a volt over potential, or less than a quarter of a volt, I should say more strictly speaking, versus off of the reversible hydrogen electrode. And almost 100% of the current density is going to see two oxygenated products. So we're making almost no hydrogen, no hydrocarbons, and these three oxygenated molecules are ethanol, acetate, and acetaldehyde. So this is a way of just saying just steering this morphology of the material can completely change the selectivity. And then we took it one step further and we started making copper silver surface alloys for the reason of really trying to change the properties of binding of intermediates on the surface. And so just you can just with just looking at the colors of these diagrams or looking as a function of voltage, these are bar charts that give you a sense of the selectivity of the different products being made. This is pure copper, this is just dosing a little bit of silver on the surface, and you can see just by eye it's striking, you're getting a very different set of product distribution really focused on acetaldehyde. And we have DFT calculations that support why the difference in chemistry can be so great just by doping the surface with just a little bit of silver. We applied this methodology to a copper silver nanoflower morphology, the high surface area, and when you put it all together, you basically get an electrode that steers about 75% of the electricity going to one specific C2 plus oxygenate, which is acetaldehyde. So this is just a way of showing by engineering the surface of the catalyst, by engineering the morphology of the catalyst, you can actually get through this complex diagram to predominantly a single C2 plus oxygenated species, but we still have a long ways to go. Now if I could have 90 seconds, I will just wrap it up with some thoughts on ammonia. Ammonia again, I have the great pleasure of working with such a fantastic team of scholars, not just at Suncat, Slack, and Stanford, but also at the Technical University of Denmark, where we partnered together on this front. And we already talked about the Haber-Bosch process, so no need to belabor this point. But I do want to make one distinction that or one addition to the commentary that I made earlier. And that is as wonderful as this process is. Once the way it's put into the ground, the fertilizer, more than half of that fertilizer ends up as runoff and an environmental problem, less than half actually ends up in the crop that you're trying to grow. And this whole distribution issue is a problem because the Haber-Bosch process requires such high temperature and pressure specifically, it really only makes sense to do it in a centralized facility. It's very tough to decentralized. So what we've been working on is trying to come up with decentralized pathways to do this, where you can imagine a solar cell just feeding electricity when the sun is shining, when there's water present, and of course, N2 is always present in our atmosphere. Can we just do this chemistry just in time and do the chemical transformation and give the crop the ammonia it needs when it needs it? And so I'm not going to go through these slides, or I would encourage you to look through some of these references. I know that we're running short on time. I just want to say that this is the board. This explains the fundamental challenge in this field is how do you selectively reduce N2 in an aqueous environment or even a non-aqueous environment without making hydrogen? Again, it comes down to a selectivity issue. I'll wrap up by showing really the solution or a solution to this problem, and that is displacing the water from the electrochemistry, so a non-aqueous system, which I think speaks well to a lot of the community that I'm speaking to today. So if you give me renewable electricity that's inexpensive, I can use that to say plate lithium metal from lithium hydroxide. I can pull out that lithium, expose it to nitrogen. It will form a lithium nitride spontaneously. I can then dunk it in water. It'll hydrolyze to form lithium hydroxide again and release ammonia, and then you can take that and cycle it back through over and over and over again. And we've shown 88 percent of the electrons actually go in to ammonia. So this is a way of, it's not a catalytic process, a cyclic process, but a way of integrating renewables into the chemical bonds of ammonia. And if you're wondering how much of your agricultural field, how much solar power do you need to drive this, how much you have to cover your agricultural field with, it's tiny. If you had an 88 percent Faraday efficiency process, you would only need five square meters of solar cell to cover one hectare worth of farmland, which is 10,000 square meters. The last thing I'll say and then I'll conclude is that if anyone's interested in ammonia and producing ammonia, one of the challenges that because it's such a ubiquitous molecule, it is very easy to pick up ammonia signals coming from the atmosphere around us. And so very rigorous protocols are needed to be able to identify that the N2 that you're feeding in is what's leading to the NH3 that's coming out the reactor. And so I just encourage you to take a look at some of these references as we've been working on some rigorous analytical chemistry methods to really be sure about exactly that question. So with that, I will conclude a simple statement. I think that electricity to X is a very promising area for energy storage where X equals fuels and chemicals. It starts with catalyst design and development. And then of course needs to get into more applied spaces with commercial grid technology to drive these chemistries at the right activity, selectivity and durability for commercial applications. So thanks again to the wonderful team of collaborators and our funding sources that have provided for all the research that you saw today. And at this point, I'll conclude and leave you with this and happy to take any questions. Thank you all. Thank you very much, Tom, for the great talk. This is a very nice way to kick off our X, the young just battery. This is the field. Thank you so much. So there are a number of questions coming to me. I think they're all very interesting questions. The first one is related to the temperature effect on the hydrogen electrolyser. So the question from the audience saying, is this running at room temperature? I mean, I'm also having a question. Is there a benefit? Let's do it at high temperature. How do you think about the temperature effect? Yeah, the temperature effect. Are we talking specifically about ammonia? I'm sorry, Yi, could you? It's for the hydrogen electrolyser. The thing about the temperature effect. Can we do a room temperature just entirely? I actually argue is maybe it's better to do a high temperature. It's more efficient. The thermal budget needs to be considered. Yeah, that's exactly right. I think temperature, so that we should definitely use that as a knob in our electrochemistry. So standard thermal chemistry, of course, temperature and pressures would drive those chemistries. And some of the chemistries can be done at under mild conditions close to STP electrochemically. And then the question is, what happens if you add some more temperature or add some more pressure? And those can have some very interesting effects. And so the way I look at, so first of all, we should absolutely keep that in play. And thankfully, there are electrochemical technologies that do operate at different range of temperatures and pressures. One of the way I look at electrocatalysis in general is electrocatalysis is no different than conventional catalysis in that you have chemical steps involved as well. But then you have now some steps that can be driven by the electric field. And so it's really a way of saying we have another kind of knob, another force we can bring to the equation. And if you look at say carbon dioxide reduction to make any one of those particular products, some of those steps might be thermal catalysis. And some of those steps might be electron driven catalysis. And so one way you can steer selectivities, if you know which ones are which, then say temperature might affect the chemical steps, the electric field will will impact the electrochemical steps. And then you can balance those in all the right ways, hopefully, to be able to steer the chemistry to where you want to go. So I think generally speaking, temperature is a very important parameter in all of this. All the chemistries that, well, I should say the carbon dioxide chemistries and the hydrogen chemistries that I showed today are STP or close to STP, I should say. You know, you can get these things to run no problem. If you crank up the pressure, no problem. If you crank up the temperature, no problem. And the question is, what's the trade off there? For the ammonia, I showed you a high temperature, but at atmospheric pressure, which is decentralizable. If you go to higher pressures there too, you get another really big set of advantages naturally, because you're trying to do chemistries with molecular N2. But we have also developed processes, I should say, that are based on, you know, room temperature and room pressure. So I hope that answers the question. Yeah, that's perfect. That's perfect. Yeah, Tom. The second question, there's certainly a lot of audience here, you know, there's some people in Zoom, there's a lot more people outside of Zoom watching. They are wondering about these questions. Hydrogen fuel cells versus the batteries for the whole world of electrical vehicles, what's your thought on that? And, you know, just of course, different type of cars on the road. You know, many a powerful lithium ion batteries now. So how does the fuel cells, hydrogen fuel cells fit into the whole picture? Yeah, I'm a big fan of batteries. And I'm also a big fan of fuel cells. And I think that this, you know, it's really interesting that for reasons that are unknown to me, right, I don't fully comprehend. People want to pit technologies against one another, or it makes no sense. The goal is sustainability. The goal is we need lots of options. You know, the fossil industry I've showed you is magnificent. We can deliver all these products. We can drive cars, buses, trains, ships, airplanes, flights from SFO to Dubai. It really is magnificent what fossil fuels can do. And so I think we need everything at our disposal. And so when I, you know, the debate of hydrogen fuel cells, for instance, versus batteries, to me it doesn't sound terribly different than gasoline versus diesel, which we choose, which is a better technology platform, kind of what's going to win. They just have different markets that they go to. And I think, and when we all, I think a lot of us recognize the differences between those technologies, there are advantages and disadvantages to each, which means that markets will, they will segment into the proper markets when, as the time goes on. And one thing I'll point out is that really a fuel cell vehicle is an EV. It's going to leverage everything as BEVs get better, FCEVs get better, because all it is is a hybrid in my book. It's basically a battery electric vehicle that has a range extender, is the way I look at it, and can go into other markets. So we'll see what, you know, as technology develops, we'll see where it goes. I'm a big fan of both. Let's push, push, push both as far and as fast as we can, because we need it. Yeah. Good, Tom. So I was looking at the number you provided. That's just amazing, the, you know, hydrogen, right, yearly production. I write down the number 65 billion kilogram per year, and close to 10 kilogram per person per year. You lease a, you know, whole chemicals, you know, that's a really big number. So I was calculating myself, I say, well, at least in my battery industry, we try to get to one kilowatt hour production per year. I'll roughly translate that how many kilograms is that my calculation show is only five billion kilogram per year, you know, hydrogen is 65 billion kilograms. So we are far away, very far away. We're talking about only gigafactory, right? We really want to see tarot factory right there, and the Tesla is a battery day, you know, Elon Musk talked about a few tarot per hour, and then up to 10. Anything you can share with us, because you deal with chemicals so much, just some insight for the battery people to think about, learn from the chemical industry, how do we get to 10 watt hour, this five billion kilogram per year type of thinking, any, any insight to share with us? Yeah, thank you so much for sharing those numbers. Absolutely. Just look at hydrogen at the United States just to put it in context. If, if the United States wanted to say we want to be the global producer of renewable hydrogen, we're gonna, we're gonna build electrolyzers and electrify that process and make hydrogen for to fit to meet the 65 billion kilograms a year need. That basically means you would have to repurpose the entire United States of America electricity grid steered towards 100 percent towards those electrolyzers and forget your refrigerator, forget plugging in your cell phones, forget, you know, all the things that we do with electricity. It has to go into the electrolyzers, and that's just a way to say, and that's just one molecule of course. And so the fuels and chemicals industry is such a massive industry that there's going to be a lot of electricity needed to pull that off. And so I, you know, it's going to be the challenges you raise are really good points on the plus side. It means that there's a good place to put that electricity as we build out renewables. If it were overbuilding, so to speak, we do have a place to put that put that electricity. But one of the big concerns that I have is once we've maxed out the grid capacity in that direction, if you can no longer run your electrolysis processes 24 7365, that's going to really hurt the techno economics of those processes. It's really tough to make an electrolyzer that runs only say 20% of the time when the sun is shining to make that cost effective. And so, so yeah, so we do need to get there and the points you raise on scale are right on the money. And we're going to have to figure out how to, how to get to that scale with all this involved time. Yeah, good. So I have a question from real, you know, I think this is a really excellent one. It's real is asking. So there's a reversible fuel cells and electrolyzer as alternative to flow batteries. And particularly what's the state of art of, you know, reversible fuel cell electrolyzer, particularly at room temperature, the efficiency, the cause, what's the limiting factor right there? You probably thought a lot about this as well. Absolutely. It's, it's, first of all, our next speaker is an expert in that area, and will be, she'll be certainly addressing a lot of those questions, but in the high temperature world, and you can also do it in the low temperature world. We've worked in that space as well. The ultimate challenge is that when you're making bonds and breaking bonds, you do pay a price. You do have to put in that extra voltage to get that chemistry to go. And typically the chemistries, the cell, the cell potentials are more like 1.2 volts instead of say like a lithium ion battery, which would be say, you know, four volts or three and a half volts. And so what that means is if you're paying say a 300 millivolt penalty in both directions, then you're talking, you know, your round trip efficiency is going to be closer to say 50% than where you really want it to be, which is more like say 80 or 90%. And that poses a big challenge. So until we can really, so I think, you know, in terms of an efficiency perspective, I think that's one major issue. And to address that, we just, we need to get better catalysts that are good at making and breaking these bonds. So it's a great idea. We're working in that space. Fundamentally, it's fantastic. And dynamically, it looks really good. It's really about the kinetics that pose the big challenge. If we can get that to work, though, then there's no reason why you couldn't do seasonal storage. Obviously, once you make these molecules, they can last for years or decades or centuries, even hydrogen is even talked about, you can make hydrogen and put it into natural gas pipeline. And maybe 2% of the natural gas pipeline is hydrogen. And the natural gas pipeline is such a massive volume that that's actually not a small amount. And that's a place that you could even without necessarily running it back into fuel cell the other way, you can still make use of that energy, for instance, for home heating or for cooking, etc. Yeah, yeah. So we have a few more questions. Let me ask come come come down to the catalyst part. You mentioned the base condition air client electrolyzer. And then there's the PAM electrolyzer right there, the acid condition, right? Compare these two, a lot of your research, you know, some of them in the really acid condition. If I look at that, you have the paired HER or ER catalysts for base and the acid condition. So for the, do you do show us your insight about acid base condition, which one is likely in the catalyst level, likely to be generally that type of the cause to meet the need? Great. Yeah, great question. So I'll start by answering that. I'll answer that question. I'll start by looking at the technology, what differentiates the commercial alkaline versus the commercial PAM? And the commercial PAM is, of course, a membrane based system. The alkaline is a separator based system. And what that means is that the alkaline is not, there's a lot of issues in terms of, number one, building up a pressure differential on both sides of that separator. Whereas the PAM, you can run a water electrolyzer and PAM electrolyzers at 150 bar on one side. You can deliver at pressure that hydrogen. Another advantage is being able to handle startup and shutdown, which the MEA design works a lot better than the alkaline design. So those are kind of like, and there are other principles I'm not going to dive into, but when we look at the commercial tech, if we had, by the way, if we had an alkaline membrane that worked as well as napion, then that could be a game changing possibility. But right now, we don't. And so all the designs are based on the napion. Thus, they're based on they need to work in acid. And so on the catalyst side, in acid, it's a much more stringent set of requirements because it's just hard for materials to survive in acid under these conditions compared to a base. So the way I look at it, you know, there are some fundamental reasons why the reactions themselves might be easier to catalyze in one electrolyte versus the other. But I don't actually think that's the dominant issue. I think the dominant issue is the materials ability to not corrode. And in base, you just have a much, much, you have a lot more of the periodic table that you can work with there under those conditions. And thus, it's easier to achieve the catalysis you want with non precious metals. Whereas in acid, you're kind of forced with precious metals for large reason wise, because of material stability, you're just much more limited into your options. Yeah, Tom, thank you for sharing your insight with us. I think our time is up. I will leave some of the question to the panel discussion, so we can talk about that more. So we start thank you so much and talk to you in the panel and just in a little bit. Now, let me welcome Sosina to come on to the stage to share with us her insights about the fuel cells. Sosina. Okay, great. It's my pleasure to be speaking with all of you. And I think some of the things I'll have to say will really dovetail quite nicely with what you heard about from Tom. I'm going to focus very much on systems that have that involve protons. And so we're going to use these kinds of systems really for this reversible storage and power generation, something that also came up in the Q&A. So the way that I envision this sort of system being included in our energy cycle is the idea that, you know, as you well know, solar and wind are coming down in price. We heard about that. And many of you are of course interested in storage. And the idea is indeed we have an electricity grid that already exists. We make use of that. And we deliver that in some somewhat local way to a reversible cell that can create hydrogen as we need it. And in order to store that electricity and can be turned on and off so that we can turn that fuel that hydrogen back into electricity when we need it for power generation. And what's nice is that you produce water, which is also very easy to store, and then use that again when we have access electricity. The nice thing about this kind of design compared to a battery is that you are able to decouple the storage and the device that is doing the electrochemical steps. So this storage container could be huge. And as Tom very much alluded to, this hydrogen could be used for anything else. It doesn't have to stay here locally. So you could put it into your fuel cell vehicle. So this is the type of device that we've really focused on in the last few years. So just to remind you then quickly how a fuel cell works. Here we have an electrolyte that transports only protons. We have an anode on the cathode on either side. We bring in our fuel and our oxygen. Normally we're using air so it's not pure oxygen, but that's the only thing that's reacting here. The two would like to get together, lower their Gibbs energy and make water. They can't do it directly because there's this electrolyte in the way. And so the reaction proceeds by the hydrogen breaking up into protons and electrons. And now those protons can go through the electrolyte. The electrons go through whatever we're trying to power. They meet up on the other side together with the oxygen and give us this oxygen reduction reaction. So that's the overall process by which a fuel cell works. And in terms of electrolysis, we're simply trying to reverse that. The way that we characterize fuel cell performance is to look at the voltage as a function of the current density. So how much current are we passing through that membrane that I showed you? Basically that electrolyte layer. And we are normalizing it per area because that's a way to be able to compare different cells to one another. So we start off at the Nernst potential. Ideally we might have some leaks and crossovers. But as we start to pull current from our cell, we start to lose some voltage. And that's due to a number of things. Here I'm just showing, in fact, a little bit of impact of fuel crossover. If you're a membrane, we're a little bit leaky. You have some finite reaction kinetics that causes the voltage to decrease because of these poor catalysis rates, some things that Tom hinted at. Then we start to see the impact of the electrode resistance. So a linear portion here, an ohmic contribution to the cell voltage drop. And then if we're trying to draw very large amounts of current out of our system, we might start to see some mass transfer resistance, which means that the gases can't get to the reaction sites as quickly as we'd like to pull them across. So ultimately, then, we define our voltage as the open circuit voltage minus all of these overpotential terms. So in some ways, similar to a battery, except that you're not looking at current density here, you might be plotting state of charge because you have a fixed amount of material that's reacting in a battery. In a fuel cell, you constantly supply the fuel so you're not running out of reacting. Then the next thing that we'd like to characterize is the power density. And so that's simply the voltage multiplied by the current density. And so we get some plot looks like this. Typically, we operate our fuel cell, not at maximum power and because that would give us relatively low efficiency. So we're about here. So we get a reasonable voltage, maybe around 0.7 volts. This is usually around a little over one volt for the open circuit voltage. And the further we are from open circuit, the lower the efficiency because the larger the overpotentials. So again, we operate typically at around 0.7 volts, which would be lower than the peak power. One way that we describe our fuel cells, however, is always by this peak power density because that's an easy single metric to compare fuel cells to one another. Now, again, electrolysis is simply doing that in reverse. And so if we have an acid electrolyte, as I'm showing here, although as Tom mentioned, usually if you have a liquid electrolyte, it's typically alkaline, but the principles are similar. And this connects then to the polymer electrolyte membrane systems that Tom also mentioned. What we're doing here is providing the electricity, driving the hydrogen evolution here by requiring the protons to react to the electrons here. That's the only way we're going to get current through this system. And then at the anode side here, we're evolving oxygen. So this is just the exact opposite of our fuel cell. And again, using a liquid electrolyte, because the oxygen and hydrogen are produced as gases, we might introduce a diaphragm here to keep the two chambers separated from one another. If we consider now what happens if we are able to raise the temperature, which is something that a question that came up, then you have a number of advantages. And it has to do with the fact, for example, that you don't have to develop bubbles using a liquid electrolyte. So if you're at higher temperature, normally you would be using a solid state electrolyte and bringing in your steam as a gas, as opposed to bringing it in as H2O liquid. And so that inherently means that you are not required to create bubbles. You're not decomposing your electrolyte in the solid state system as you are in the liquid system. So here you have to actually decompose the electrolyte itself, which is a liquid water. Here you're bringing in hydrogen gas, steam gas, breaking it up, creating oxygen and hydrogen, which are also gaseous. And in fact, that means that they're naturally separated because you have the solid electrolyte. And it also means that you can work with lower voltages, which you may have from your PV cell, for example. Okay. And of course, the catalysis is much improved. So now let's combine the way these two devices look. So here's our electrochemical cell. Let's call it fuel electrode and air electrode because anode and cathode change definitions depending on where we are, which direction the current is flowing. So we'll be safe by calling them fuel and air electrodes. This is what I showed you before for the fuel cell mode. And now if we simply reverse the current, we're in electrolysis mode. So that would be our curve. This is again, typically around one volt. And the Gibbs energy for the reaction, this is just going to your physical chemistry. We've got the standard Gibbs energy. And then it depends on what partial pressures we're bringing in all of these reactants. Okay. Our Gibbs energy is also, we can write it as delta H minus T delta S that also it defines for us our Nernst potential. So that's what would be right here if we have these various gases on either side. And again, we pull current and we get power and fuel cell mode. If we pump this up compared to where we were, we're increasing that voltage and we're doing electrolysis mode. One thing that, and in this case, the overpotentials are that difference in electrolysis mode similar to what we had in fuel cell mode. So one thing that makes the electrolysis mode interesting is that this gives energy, we can break it up into the enthalpy contribution and the entropy contribution. And so let's say that if we now compute the voltage associated with that, let's put it right here. And what's interesting about that is that this becomes the heat requirement in order to operate this reaction because the this reaction electrolysis requires heat in. So this is an endothermal reaction. So the amount of heat that we get from the inefficiencies of our system are defined by this overpotential. And if we're have a small over potential such that we're below this voltage that corresponds to the enthalpy, we actually have to input energy into our system, a thermal energy into our system. On the other hand, if we operate at some higher voltage values, then we actually have heat released because our inefficiencies are so high. But that's actually not bad because we're using that heat to actually drive the the delta H need for this system. Now, oftentimes, we operate at what's called the thermal neutral point right here, such that our inefficiencies in catalysis exactly balance out the needs that we have for the heat input to actually run the reaction. Another thing that you'll hear a lot in terms of these fuel cells so that I forgot to say this is around 1.3 volts. So that's the thermal neutral voltage. And that's the current density that we'll specify as a metric for how good our electrolysis is, our electrolyser cell is. Another point here is that generally increasing the temperature lowers the electrical demand, but increases the thermal demand because you have to put in heat to increase the temperature. But again, if you have inefficiencies that sort of takes care of the heat load, and or if you have some other sources of thermal energy, you can take care of the heat load. So there's a lot of arguments made about why high temperature electrolysis should be more efficient than lower temperature electrolysis. And then just to remind you fuel cell mode that would be our power density curve. And the two metrics then of interest are the peak power density of our fuel cell and the current density in electrolysis mode of our electrolyser cell. Okay. So there's a push towards intermediate temperature at sort of the race to the middle. That's all puns intended. So what we have here are the conductivities of a number of different materials. And let's say we want to operate at somewhere between around 250 and maybe 550 degrees Celsius. And we have here material dope barium zirconate, which has a very high conductivity in this range. There's also dope Syria, which is a fluoride structured material, which has very high conductivity as well. I'll just show you those structures. The Syria material here has oxygen vacancies. That's why it's a very good oxygen ion conductor. The dope barium zirconate is a material which has a perovskite structure. And we dope that such that it has some sort of prevalent and that is charge balance by having protons incorporated into the structure. They make these hydrogen bonds and they can move around through the structure. And that gives you this very high conductivity. So I'll show you what's been done with this intermediate temperature electrolyte system because as I said, race to the middle gives us a lot of the benefits of both room temperature operation, which is easy on our cycling. No need to have very high temperature resistant materials, but much better catalysis rates than room temperature operation. And also, as I said, to some extent, better electrochemical efficiency for electrolysis. Now I described to you also this oxygen ion conductor for electrolysis. Well, let me tell you some differences between operating a fuel cell using a proton conductor versus using an oxygen ion conductor. One of the key issues is that in fuel cell mode, once you provide the hydrogen and moves across the electrolyte, the product water is produced on the air electrode side. And so there's no dilution of this hydrogen as you run the process. In the oxygen ion conducting system, there is dilution because the oxygen ions are coming across reacting with the protons and we make water. So if you have a stack, then the first cell will have a high hydrogen concentration. But then as you go through the stack, you get less and less hydrogen. So the efficiency and sort of system design is not ideal. So that's one difference. On the other hand, if we're using these ceramic electrolytes, the construction is quite similar between them. So this would be a thin cathode, a thin electrolyte supported on a thick anode. And again, I'm going to call this now the fuel electrode and this the air electrode because we're going to start talking about electrolysis. And this is going to matter in terms of electrolysis, which one is the thick electrode versus which one is the thin electrode. So let's look at the behavior in electrolysis mode. What happens is that, and I should also point out here that the fuel electrode has is a mixture of nickel and the electrolyte material. So this we can make as a big component. The hydrogen electro catalysis is very fast. So we can allow that to be the support structure. So in the electrolysis mode, we have, again, we bring in water to the air electrode. And what's nice here is that this is very thin, so you don't have mass diffusion losses. The other aspect is that the nickel, the fuel electrode is on this side. So the nickel is never exposed to steam. In the case of the oxygen I'm conducting electrolyte, the nickel is exposed to steam. And it's actually exposed to a lot of steam because we have to have a lot of steam here in order to not have mass diffusion losses to get this sort of bulky molecule all the way through this thick electrode. In the case on the proton conducting side, hydrogen can easily make its way out here. And this is a thin electrode. So the water and the oxygen, which are slightly bigger molecules, can make their way through this thin air electrode. And then in this case here, you'd have to separate the fuel afterwards from the water. It's not so dramatic, but it is a disadvantage. From the electro catalysis perspective, this is much better because hydrogen evolution is really very easy. Here you have to do oxygen ions coming out and becoming neutral oxygen species at the air electrode. And to show you what that looks like, the consequence of that in the oxygen ion conducting system, because this over potential is large, you build up a very high oxygen, effective oxygen partial pressure in the electrolyte. And at this interface, you start to create some oxygen bubbles, even in the electrolyte, but certainly at that interface. So this has been a real challenge to have high stability for the oxygen ion conducting systems. So if we look at the proton conducting system, we've talked a little bit already about power generation, so using hydrogen to make fuel to make electricity. And we've talked about it running that in reverse to make a hydrogen. But there are all sorts of other things just sort of alluding to the idea of making chemicals with these electrochemical systems. You can do that certainly with these solid state systems as well. I'm just going to say a couple of words on the CO2 reduction, on the nitrogen reduction as well, and sort of how these work well or not so well. And again, let me do the comparison with the oxygen ion conducting system. In the case of the proton conducting system, you bring in ideally you bring in water, the protons come across, and you react to CO2. And ideally you're going to make some mixture of CO and methane. And again, we don't want to have the hydrogen evolved here. As Tom alluded to, this is a very easy reaction, this hydrogen evolution. So it's actually quite hard to prevent this reaction from occurring. So if you're interested in CO2 reduction, it actually makes a lot of sense to use the oxygen ion conductor. Because here, you simply pull off oxygen from the CO2. The reaction that you're competing against is carbon deposition, which is not terribly favored. So this actually goes quite well for CO2 reduction. If you want to make more interesting molecules, that becomes a challenge. But for CO production, this is very straightforward. In the case of nitrogen reduction, you have this challenge again of hydrogen evolution. But it is one that I think is sort of a holy grail and one that is really worthwhile to pursue. In the case of oxygen ion conductor, you really wouldn't try to do this because you have to pull oxygen away from this water molecule and then convince that hydrogen to react with nitrogen in a way that doesn't allow just hydrogen to evolve. So this is pretty unlikely. This is actually almost commercial level already. And in the case of making ammonia, the proton conducting system has, in my view, really a big advantage over the oxygen ion conducting system. You also would not want to go very high temperature in this case. The temperature question was raised. You'd not want to go to high temperature because nitrogen and hydrogen thermodynamically will dissociate. So thermodynamics are in your favor at lower temperature. You just need high enough temperature in order to catalyze the reaction, get over those reaction barriers. So let me just tell you some of the progress now using these proton conducting oxide electrolytes. So in about 2014, when we started looking at this and saying, you know, we should get involved in this because these electrolytes are really very good. I've just described to you all their features and why they ought to work. In the state of the art and fuel cells in terms of that peak power density at 500 C was really quite low, despite the fact that the proton conductors have higher ionic conductivity than the oxygen ion conductors. So peak power density at 500 C was much, much higher in 2005-ish than we were getting from the oxygen ion conducting systems. In electrolysis, there were some places where it was somewhat better, but still not great. So and there was generally somewhat limited research activity in this area. So we went ahead and first of all tried to understand what are the factors that are limiting the performance. One of them was achieving this high conductivity as well as good stability in the electrolyte. Another was poor contact with the air electrode. The air electrode has poor electrochemical activity and difficulty to make the cell because this thin electrolyte is actually challenged because these proton conducting oxides are rather brittle, require high temperatures to process and so on. So generally a processing challenge. And so we set about to tackle this first of all for just making some better electrolytes and the way that we did that had to do with modifying the chemistry just a little bit. And when you modify the chemistry, what happens is that you're able to get some very large grain materials. So this is our polycrystalline electrolyte and it's the grain boundaries that are very resistive in these materials. So if you have a large grain material, the number of these barriers that you have to cross is diminished compared to this fine grain material. This is very refractory and very difficult to work with. And again these small grains means that you have lots of these barriers to go across. So we developed this composition. But it has enough zirconium in it to prevent it from reacting with CO2. So a competing composition has about 70% cerium in it reacts very quickly with CO2. Whereas in the case where you have just 40% cerium, no reaction with CO2 at any temperature of interest. So that was the combination electrolyte composition we developed. Then looking at the poor electrochemical activity of these electrode materials. So if we look at what makes a good arrow electrode or good cathode for the solid oxide, traditional solid oxide fuel cell systems, this is supposed to be saying an arrow that goes forward and reverse. What we have ideally is a system that does electro catalysis on its entire surface as opposed to being limited to just this what's called a triple phase boundary. So we want something that can allow oxygen ions through it as well as electrons to it. So we call that a mixed oxygen ion electron conductor. And if it allows the oxygen ions and the electrons to go through, that means we can ideally have catalysis all the way across the surface and we're not limited to what people call these triple phase boundaries. In the case of a proton conducting system, then we'd want something which is a mixed proton and electron conductor. Again, this should be an arrow going forward and reverse, allowing us to do both electrolysis mode as well as fuel cell mode. Well, that would be the case what we want. But in general, what people have used is this exact same material as we use in the oxygen ion conducting system. And there's some fear in fact that you might make water right here. If you use that material, the protons would come in, the oxygen ions would come in, the electrons would come in, and you'd make water at this interface would be very very bad news. But you might try it and hope for the best. And what it turns out is that many of these materials that we viewed as mixed oxygen ion and electron conductors are actually triple conductors, meaning that they in fact do let the proton through. They let water molecules do them as well. So they're kind of squishy for all sorts of ions, the oxygen ions, the protons, the water molecules, the electrons. So everything can go through them. And that's sort of our lucky break that they in fact work quite well in terms of the transport characteristics. Then we need the catalysis characteristics to be very good as well. And so the material that we've worked with is this large fraction of the periodic table. So we just call it PBSEF. And we first established that it was chemically compatible with the other components. And then we established that indeed it will uptake a lot of protons when we expose it to some small amount of water partial pressure. So again, sort of squishy to oxygen ions to protons to water molecules, all of those can go through. And then we separately measured the electronic conductivity. So it's a good electronic conductor as well. And then we did an experiment where we varied the area over which this electrical chemical catalysis can occur to establish that really are the protons going through and can we get a high activity. And so when we did that, we saw, yes, the activity scales with the diameter or with the area here. And because it scales with the area, that means the entire region is electrochemically active. And so with that, we're not limited to the triple phase boundaries, great stuff. So then we said, all right, let's deal with cell fabrication. We have this really poor contact with the electrolyte, which is increasing the electrode resistance cell resistance. And we did that by depositing a thin layer by pulse laser deposition of this air electrode material. And that allows us to have good contact with the overall porous cathode. And this would not make any sense unless we knew that this layer here would be permeable to all the species of interest. And after doing that, we're able to get record power generation, so electricity generation mode, hydrogen production on this side here, electrolysis mode, works very well in both cases, quite stable, even with that PLD layer. And so if we now we can also run it in reverse. So indeed, we can do hydrogen storage and electricity production back and forth. And again, we have this nice operability as a battery. And if I just put that show you, you know, what's the impact of putting in this thin layer? What's the impact of comparing with some other state of the art cathode materials? The overall we do better both because we have that dense layer and because we're using a high activity cathode. That's true. The impact is true for both electricity production and for hydrogen production. So just to put it in the context of where the community stands. So we published this in Nature Energy and we had the record for about a month. So this is our results and it was nicely covered on news and views. But then about a month later, this new result came out and another few months later, this high power density also came out. So it's just to say that, you know, with engineering effort and realizing that this is important, we can we can really take advantage of material properties for hydrogen production. We are still highest current density that's been reported. But what I'm showing you here is that in fact we realized there's a little bit of a fly in the ointment, which is that we have electronic leakage. So if we look at just the proton at current, that would be about here. Everything beyond that is in fact electronic current. And this has been really not recognized in the community. So this is an area that we're really working on currently. So with that, I want to just thank my group. I like to show this picture with Chicago Skyline in the background because it's very lovely. But in fact, we know that the way we really handle our groups now is through Zoom. And again, I just want to thank these great people who've done this amazing work and continuing to work on today. And thank you for your attention. Susanna, thank you very much for that wonderful talk. And I apologize to you and the audience that have some rowdy kids in the background. So let me kick off with a hard question so then I can mute myself very quickly. Both you and Tom talk about ammonia as an attractive hydrogen carrier. Is it also an attractive carrier when you're using it for energy storage in a reversible context? I would certainly think so. I mean, there is increasing work to use ammonia directly in fuel cells. So that certainly has happened and is continuing to happen. We've been looking, RPE really made a push for including ammonia in the energy infrastructure. So I would say that they get a lot of credit for that because the reason it comes up is because hydrogen in the sort of pipelines is really costly. And this is why the U.S. backed away from that and backed away, frankly, from fuel cell vehicles as well because the hydrogen pipeline was going to be just incredibly expensive. And so ammonia can be made as a liquid fuel. But I wouldn't put it in a car. It's just not the toxicity is an issue. So you can use that ammonia either, as you said, sort of at a remote location where you've got solar electricity and you want to store that electricity for later use. Certainly you can use a fuel cell in that mode. Again, that activity is just starting. You can also use an electrochemical cell at that site to convert that ammonia to hydrogen. So what's happening on the other electrode when you're oxidizing the ammonia, what's happening on the other side could be hydrogen evolution, or it could be reacting that hydrogen with oxygen to get electricity out. So you have a lot of flexibility there. So, to follow up on that, how is the reversibility of the ammonia half reaction? Well, I wouldn't, we haven't tried that yet, I would say. So what we've done, the challenge comes about where it's really solving the ammonia production half of it. So using ammonia in the fuel cell mode is relatively easy. There have been papers on that and we've done the electrochemical hydrogen pumping out of that. But getting hydrogen to go across the electrolyte and actually react with the nitrogen, that is the killer. And that, you know, and Tom said, you know, let's just do this thermochemical cycle instead. That is certainly one that requires a lot of work. And if we come up with a good catalyst for that, you know, then you run into, is that really the right catalyst for doing the ammonia oxidation? So for the PEM electrolysis cells, for example, you don't use the same catalyst for fuel cell mode and electrolysis mode. For the higher temperature systems, you can just as I showed you. So that question is still really an unknown because we don't really have the good catalyst yet, even to make the ammonia. So then asking them to also be reversible, that's, you know, that's further down the line. Great. Well, let me get both you and Tom to think about for the panel discussion on the importance of round-trip efficiency. I think this is kind of a big question mark in many people's head. Susanna, let me switch gears and ask about the value of products. Right? So if I understand correctly, if you're using an oxygen ion conductor, you're generating pure oxygen on one side, which there's no need to store, we can just get it back in air. Is it a significant value from the operation of a battery plant, right, with a solid oxide fuel cell as the main brain? If you sold at the pure oxygen, is that going to be a significant factor in offsetting the energy storage costs or that's not a big deal at the moment? So we've, you know, the gas companies periodically, we ask them this, do you, you know, we could give you a really pure oxygen by electrochemical pumping, right? So you could bring an air on one side and pump oxygen to the other side. And I forget one of the companies was actually trying to manufacture this because in laboratories, you might need a pure tank of oxygen, right? And you can just get it out of the air. So oxygen generators in the same way that we have hydrogen generators that use, you know, waters, the input and, you know, in the laboratory, we get pure hydrogen out of that. You know, the economics were not quite there to do that. So I don't pretend to be a techno economic analysis person. I know that, you know, from combustion perspective, there are times where you'd like to have pure oxygen rather than air. Also, if you actually think about sequestering the CO2, if you had pure oxygen delivered, then you don't have to do the subsequent separation between a CO2 and nitrogen so that you could put your CO2 in the ground. So these are areas where oxygen could be valuable. But I can't say for sure that that's going to be the driver for that one. Thank you. So I think now we can get into some material science questions. There are a number of questions from the audience. So remember that a lot of our audience are from the battery lithium-ion battery field. So I'm seeing actually a lot of intersection in the materials questions between batteries and fuel cell. So let me maybe draw your attention to solid state batteries, which is looking remarkably similar to a fuel cell. And I think something that is quite shared is the solid electrolyte-electrode interface. You hinted in your talk on the importance of this interface, you talked about modifying it. Can you say a bit more about how limiting this interface is? You know, one typically think it's the catalytic interface that could be limiting, but I think you are also implying that the solid-solid interface could be limiting as well. Right. I mean, so they're in some ways in the high-temperature gas system. It's actually a little bit easier than the battery system because your electrodes are generally not changing volume. They're pretty stationary when you operate. So as long as you get a good microstructure when you start and you have something that doesn't evolve very much over time, then you're good to go. But you do have to understand that there will be some sort of electrochemical reaction at that interface. In particular, as we go towards more and more of these mixed conductors where you actually do have to have ions move through the electrode and then at that electrode-electrode interface make their way over. And so we have to make sure that that's not the rate limiting step. In some instances it is. Usually it's still not the rate limiting step. In most cases, the rate limiting step still is the reaction at the surface of that mixed conductor. Sometimes electron transport, depending on your design, whether or not you have another component in your electrode structure because oftentimes, just like in the battery system, we don't have sufficient ionic conductivity in our electrode material. So we have a second component or sometimes we don't have enough oxygen ion conductivity. So we have a second component which is there simply to bring the reactant, either the ions or electrons in. And then, again, we have to make sure what the contact is between those and make sure that we have enough of them. In the case of our system that we looked at, the challenge was really that the cathode or the air electrode material had a very low melt temperature, whereas our electrolyte material needs to be centered at 1600 C. And so we just couldn't get to a high enough temperature to have good contact between that air electrode and the electrolyte. And so putting an appeal delay or make a difference. But absolutely, we have to be looking at those interfaces more. And there are instances where an electrolyte may not be stable. And so under reducing or oxidizing conditions that happens in different systems that we're looking at, where we want to do a lot of the types of things that batteries people do, which is put in a quote unquote SEI, meaning a layer that will protect the electrolyte. And even in our mix, in our protonic system here that I talked about, one of the ways to try to get rid of that electronic conductivity would be to put a component which is a pure proton conductor. And even though it has very low protonic conductivity, if it's thin enough, it's not going to be a barrier. Great, Susana. So there's another related question. So in the battery field, there's a lot of effort at increasing the contact area between the electrode and the electrolyte, which is easy to do when it's a liquid. The example you showed was a flat coating on a flat solid electrolyte. Are there opportunities for sort of 3D nanostructuring or microstructuring just to enlarge the surface area, for example, in a composite electrode? Is this something that is worthy of investigation? Yeah, so just let me clarify that the contact that we made, there that it was most certainly not a flat electrode. It was a 10 micron thick porous electrode with as much surface area as we could get and maintain at that high temperature. The reason that we had that thin layer is simply to have better contact between the air electrode material and the electrolyte. That was all that that was doing, was not the place for the electric catalysis. I mean, it has some electric catalysis there, but the majority of it is happening on that 3D porous electrode. The challenge overall for the high temperature systems is that you make a beautiful 3D nanostructured electrode, and after 10 minutes of operation, everything is coarsened away. So absolutely, it matters, and we try to make such structures that we always have to be careful that they can last. And that is also sort of the driver to go towards these intermediate temperatures, sort of the race to the middle, as I call it, because then you can take advantage of those structures and use it. Susanna, thank you so much. It makes me want to work on solid oxide field cell even more so today. So let me bring Tom back and EE, and we can have about 20 minutes for a panel discussion. All right, we'll come back, Tom. So maybe let me take the liberty to kick off with the first question, and it's going to be a provocative one. In both of your talks today, you really emphasize the need for high current density, the need for high round-trip efficiency. And the arguments, of course, are obvious. You have lower cost of operation, if you have higher round-trip efficiency, and if you have a high current density, then your capital cost is lower, the cell can be smaller in area. So the provocative question as this follows. In the era of inexpensive electricity, which Tom really highlighted, does round-trip efficiency matter that much? And secondly, as we look at long duration storage, so high energy to power ratio, say, for example, a week, then the cost of the power stack or the cost of the cell also decreases in the system as well. Then does the current density matter? So the provocative question is, are these two conditions, the lower cost electricity and the need for long duration storage, is fundamentally changing the balance for the commercialization prospect of fuel cell technologies for storage? Susana, do you want to start? Yeah, I mean, I would say that you're throwing us a lobel. Right? I mean, that's the argument we would like to make. The challenge is sort of convincing somebody else that, hey, don't worry about your round-trip efficiency, because you've got hydrogen that you can store for months on time, and there's no self-discharge here. But I think it's a moving target, because there will always be competing technologies that have great metrics. And so I do think that until we can show some sort of reasonable round-trip efficiency, that even if you have year-long storage, again, I'm not a big techno-economic person to make specific numbers, but I can tell you that most battery people, that's the first thing they're going to ask is, well, we have this X value of round-trip efficiency. But I should say that both the fuel cell, reversible cell and a battery, if you operate them very close to open circuit voltage and go back and forth very gently, your round-trip efficiency is going to be very high. So yeah, you're going to trade off the current density for high efficiency. So yeah, I agree that there's always going to be specific instances where you need a different set of metrics. But if you can say, I've got the slam dunk technology that will cover you for all situations, of course, the rest of the world is going to take you more seriously. Yeah, I think it's worth noting that the current density in the battery is a thousand times smaller than in a fuel cell. It's really not a fair comparison when you compare round-trip efficiency between the two. Tom, it'd be great to get your thoughts as well. Yeah, great. I mean, Sosina made some fantastic points that I'll just add that this is why techno economics is going to be really important to everything that we do moving ahead, because every type of technology, every market that's out there has its own set of battery conditions and scenarios to play out. This is where it's very helpful to do some just pen and paper analysis up front can steer us in some interesting directions. But at the end of the day, to do a proper techno economic analysis really requires some dedicated effort. And it's not always easy to do that with technologies that effectively don't exist today. It's a lot easier when you're doing a techno economic analysis on a methanol synthesis plant, where there's already 100 of them in full blown operation today, and you're just making the nth plant over 100 very different scenario than the type of technology we're working on. So the points you made, Will, are fantastic. I think that is true that basically the landscape around us is changing. And so that's why technologies that might not make sense at this moment could make a lot of sense given another decade or two given the trends that are going on. So I'm not a hockey player, but I've learned that you don't skate to where the puck is, you skate to where the puck is going to be. So maybe Will, I'll do the next question. Fantastic talk, Sosina. I think Will asked you a little bit of that already. I thought about both in your talk and also in Tom's talk. This one thing right there, Will used the term I like to use as well, 3D electrochemistry. Tom, in your case, you are dealing with this liquid, this solid catalyst, this oftentimes gas species right there, three phase. And Sosina's case is a solid electrolyte, solid catalyst, and gas phase, you know, also three phase, slightly different three phase. So thinking about 3D electrochemistry, could you make some comment on that, you know, in terms of performance, the power related to this energy, run to efficiency, maybe related to the cause eventually. And the insight to share with us and also the future direction you are seeing we could potentially move into. Right. I mean, I think, so in some instances, what we've tried to do is measure the absolute value of the electrochemical activities, say on a flat surface, and then extrapolate from that what type of 3D microstructure we could get that could be stable, that could be the last, and how much that would impact the performance. So that's a very, you know, sort of ground up. Let me try to build the best of electro that I can make. Then there's the other approach, which is I know how to make nanoparticles. I know that, you know, high surface area is valuable for me. Let me just see what I can actually make. Right. So there are two different approaches there in terms of what we do in the laboratory. And, you know, I think there's just no question that overall higher surface area, because we have a gas solid reaction is going to help the overall current density. Right. If we can do something to also increase the rate at each reaction site, that would be advantageous. Again, the challenge in the high temperature system is that you need to have that structure be stable. The other challenge is that you do have to make sure that the porosity itself, that the pore sizes are big enough that you don't get mass transport limitations. So sometimes we get so excited about nanostructuring and that we forget that the pores should not be at the nanoscale. The pores should be closer to the micron scale, whereas we still have this very high surface area. So we've tried to do some, you know, cool, fun things like do inverse opal electrodes, but they're all hard and ultimately having a high surface area. We just know that that's the answer. So how much more you get out of having it be, you know, a beautifully ordered structure, that is going to impact the gas transport. And as long as we have a reasonably porous structure, the gas transport limitations are not dominant. So yeah, the materials chemist in me would love, you know, sort of making these beautiful structures that the reality is they work quite well as long as we have high surface area and good porosity. Yeah. Tom. Yeah, thank you. I'll add a couple things there. And I think the idea here is, you know, when it comes to reinventing technology revolutionizing technology ahead, this is an opportunity where you have, it's not just about the catalyst, it's about the device design as well. And the question is, can you improve that device design for improved functionality? And can you reduce the capital expenditure? So, you know, those designs that we have today for MEAs, whether it's for fuel cells or for electrolyzers in the low temperature world, you know, that all stem from research and development from half a century ago developing those first sets of fuel cells. And now we can port that to all these other technologies, be it for hydrogen, be it for ammonia, be it for CO2. And, you know, the question is, is there a fundamentally different platform that we could design that not only achieves higher reaction rates based on the mass transport points that were made earlier, but also perhaps, you know, lower cap acts, are there simplified designs out there? And can we use those designs to do things that we might not otherwise do in a conventional format? So for instance, we've seen in CO2 electrolysis, which again is very different from hydrogen production. Hydrogen production, the selectivity is not the biggest challenge in that space. In CO2 electrolysis, it's a massive challenge. And can we use the three-dimensionality of the electrode? For instance, can we design a morphology so that you can actually execute chemistries to get you the desired product in a more selective way than you otherwise would if it were, say, just a flat electrode, even if the catalyst is the exact same catalyst? Can we use, in other words, say, mass transport principles where we know at the electrified interface, you're going to have concentration over potentials. The pH can be different, you know, different cations, different anions, you know, in one spot versus another. Can we use those types of principles to our advantage in steering chemistry in the right way? So I would encourage all of us to think about not just, you know, what's happening at that molecular scale, but what's happening in a three-dimensional continuum scale and how can we stitch models, multi-scale models in both length and time, so that we can understand that and control it to our advantage. Thank you, Tom. And so seeing that for this insight. Well, let me ask another provocative question. So something that's being on a lot of the battery folks's mind, and I think certainly a topic of excitement, is decarbonizing aviation, right? And I think Tom hinted at that as well. This is something that is very hard to decarbonize, and battery technology is going to be certainly a long way to go. And the same thing for ocean freight as well. These are huge CO2 producers, but it's going to be amongst the last one to be decarbonized. So my question has to do with, you know, what would it take to decarbonize aviation using some sort of an electoral fuel? And I sort of see two ways, right? So you can either make hydrogen work on planes, and indeed there's been announcements from Airbus on making hydrogen planes, but one can think about the challenges there too. Or you could think about using electoral fuel that's liquid, carbon containing, but then you have to think about CO2 capture at the end. So I was wondering that both of you can talk about a little bit, sort of the prospects of electoral fuel and electric flight and perhaps ocean freight. Is this enough and Tom? Yeah, I mean, of the fuels that could be used in an electrochemical sense, this is actually a place where I would think ammonia has a role to play directly in a fuel cell, because as you say, there's nothing to capture. And so that is actually reasonable. And you also, it's not like having a consumer pump liquid ammonia into their gasoline tank. So I think there's a possibility of doing it that way. And then you don't have all the challenges of keeping pressurized hydrogen and having enough of it on a plane, because all the pressurization means you have this huge, heavy tank. And so it's not clear whether you're ever going to be able to have enough fuel. So that would be my view. And as you said, there are, of course, airplane manufacturers that are looking at hydrogen based fuel cells. For ships, there was a period of time where the Navy was really looking at putting fuel cells on all of their ships. The saltwater is certainly a challenge. And in their case, and for submarines, there was a real challenge of also bringing oxygen on the vessel. So I think there's certainly room to address this. There is also the idea that you use biofuels, because that's the CO2 capture that you're doing. And if you look at how much fuel is in aviation compared to everything else we do, you could imagine that crop-based CO2 could address that need. Again, but there's expectation that once we get over our current scenario, there's going to be more and more air travel. So it's not clear that that's going to remain a possibility. So my take on it is that ammonium might really have a role to play. So I think I will let's see. Stan, are you here? We have our, you know, Stan Witingham right here, the Nobel Prize winner. He likes to ask a question. I don't know. Stan, can you turn on your microphone? Okay. Hello. Hi, Stan. Hello. So I was very curious about this. I worked on hydrogen ox and electrolysis, my bachelor's degree many, many years ago. And at that time we spend a lot of time trying to build fuel cells with bipolar technology. Is bipolar still the way to go? Well, I mean, so to make it reversible, I mean, we demonstrate that. That's much easier in the high temperature systems, because the same electrodes work in forward and reverse. Whereas again, in the liquid systems there, you know, Tom can comment on that, but generally you're not trying to, I shouldn't say, it would be great to create that. But to my understanding, there's not been a huge breakthrough on making reversible cells there. So as you say, bipolar. Yeah, I'll add for sure. I mean, people have demonstrated that technology. It exists is just not in a commercial ready state yet. So, you know, I just really, and there's no showstoppers. There's nothing that says that it's not a good idea to build out that tech and see where it goes. I mean, in fact, that's how I look at all of this, is that of course, we don't know what the future holds in the next 10, 20, 30, 40, 50 years. And I view that our role in this as an academic research lab, as a national lab, along with all our colleagues around the globe is to really provide options and to start, try to ultimately de-risk a lot of that technology and see where it goes. In some cases, we can, you know, hand it off to a larger commercial entity that has expertise in these areas or our students and postdocs found companies and start those things up and see where it goes. And so, you know, it's definitely worth investigating. There's no question in my mind that developing cells that can go backwards and forwards is a very, very good idea. The challenges are mighty. No question about that. So, but that's what we do. We have to, these are hard challenges and we have to try. We need options. So thank you very much for a very interesting session this morning. Thank you, Stan, for the great questions. So maybe I'll ask the one last question before we end today's session. Both of you touched upon a little bit already. Very important is the lifetime of these electrochemical technologies. And let me mention in the batteries field, right? People do charging, discharging cycles and this has a huge implication on the cost. You know, I would imagine if lithium-ion batteries, you know, the Grafiano MMC can run 30 years and 11,000 cycles, keep high energy density, reasonable power, that has huge impact in the whole economics. So in the technology you present today, and share with us the lifetime consideration, right? Tom, you touched upon some corrosion. So Sina, you mentioned about high temperature can change things, right? Share with us the, you know, the current status and the challenge ahead and where you can see this will go and what does it take? Yeah. So let me take this one first. I have to tell you, I mean, that's an excellent question. People ask it all the time and I get frustrated because I cannot see degradation in myself. Okay, we did 700 hours and yeah, there's a little bit of up and down, but and we look at the microstructure afterwards. This is why, you know, 5 to 600 C for ceramics is a good idea. You're not at 1000 C where you start to get reactions between components and microstructure evolution. So I'm just going to not really answer that, but tell you that for these systems, the key challenge that we face is this faraday efficiency because we get a little bit of electronic conductivity. And even though our transference number is around 96% for the oxygen ions, or sorry for the protons, in electrolysis mode, you're effectively oxidizing the electrolyte as well. And that makes it more electronically conducting. So that's where you get a penalty in faraday efficiency for electrolysis. It doesn't penalize you at all in fuel cell mode because you're effectively reducing the electrolyte itself. And that's where it becomes less electronically conducting. So that's our challenge right now. To be honest, stability, as far as we can tell, again, hundreds of hours, of course, the university laboratory, that's about as much as you can do because people have to do their PhDs and move on. So if we could get companies that have stat testing and can keep it running for 1000 hours, that'd be great. But so far we have not seen any degradation. There are other systems that have degradation. The air electrode in solid oxide fuel cells versus strontium evolution. So if anybody is in that field, I mean, they're definitely that the composition doesn't stay the same over time. That's an issue. But again, lower temperature operation addresses that. Great to know, Selsina. Great to know, Tom. Yeah, excellent. You know, definitely this a lot of this tech is already commercialized, as we were talking about before. I mean, just to give some numbers, which mean it was just commercialized naturally, that means it must have some reasonable durability. But the challenges as you were pointing out is just because things might be durable for the typical application of a battery in a laptop or a battery in a car, that's a different type of durability than one might need if one wants to plug energy storage technologies into wind bills and solar cells and have them last for decades. So just to give you some numbers, some food for thought. So fuel cells, which a lot of this tech is kind of based off of, as I mentioned before, with membrane electrode assemblies and gas diffusion electrodes that construct. So these days, low temperature fuel cells are typically last about 5000 hours of operation. And 5000 hours is, if you think about it, that's a new average say 30 miles an hour driving. That's like 150,000 miles of driving, which is a reasonable amount that somebody might drive before they're looking to upgrade their car. So just to give you some numbers there, of course, you can always replace the stack itself. The stack itself is not the dominant cost. You know, if you're producing large scale manufacturing of fuel cell vehicles. Now 5000 hours, there's two things I want to say about that number. Number one, there's about 9000 hours in a year. So when you integrate, you know, it's only half a year worth of use. And that doesn't sound like a whole lot if you're doing chemical manufacturing. That's now on the plus side. That is the most brutal drive cycle. I mean, the drive cycles, I should say, are the most brutal like variability you could possibly imagine. And just think of all of this when we're driving our vehicles around, like the starts and the stops and the accelerations, decelerations, you know, that can't be any worse than a solar cell, I would think, in terms of, you know, really a challenging environment for it to be in. So that's very good that at least it's able to handle a very heavy variable input or I should say load to operate. So that's good news. Now that check was translated into the PEM electrolysis world. And the PEM electrolyzers are sold these days. A typical warranty is like seven year warranty. So seven times, you know, 9000 hours a year. That's over 60,000 hours of use. That's the warranty. It doesn't mean it's going to fail after seven years. It's kind of like, you know, the way that they would say it is like buying a car with a seven year warranty. It's not like you expect the thing to fall apart at seven years in one day. Hopefully maybe it can last two times longer, even three times longer, longer. But that's what that's the warranty that they provide. Now the electrolyzers, all electrolyzers today really are meant for 24, 7365 operation. There's a lot of research ongoing at this moment to really understand, and that's still a to be determined, what happens to a PEM electrolyzer when you're giving it a wind turbine's worth of cycles or solar cycles. And so the good news is that it's way better than alkaline could ever hope to be, and it's performing reasonably well. But you know, what is that, you know, there's still a lot of studies need to be done in that space. And so that's just to give you, you know, so these are really important durability metrics that need to be paid attention to the technology is performing reasonably well, given what it's being asked to do. And naturally, we need to ask ourselves the question, how would this perform differently? If we're applying that technology in a different sector, a different space, and so still lots of great research questions to ask. And I think partnering with industry is a great way to go about doing that. I really completely resonate with you that I think the use cases we have today are very different than what people have thought of, say 20 years ago. And I think stores actually create some very interesting use cases that pivot the needle a little bit for technology you mentioned that haven't reached commercial scale. You know, I love having metrics. But I think sometime if we look at the Department of Energy Metrics on fuel cell, they're not really designed with energy storage in mind, they're designed for making electricity 24 seven. So I think for all the researchers out there, I think it's really worthwhile to think about the economics. And you know, maybe you don't have to have a very low over potential, maybe you want to have high over potential and long lifetime. And that may sort of allow us to have extra not to tune. So Tom and Sissina, thank you for making that very important point. I think with that, this is just an amazing session today. Thank you, Sissina and Tom for sharing with us your insight. So at last, Justin, can you bring up our next event slide? Next event is two weeks from now, October 16. We are going to have our industry panel, you know, today we touch upon how do we get to tell about our scale, right? The storage acts and the next event will be talking really look into how do we get there and this really, really big scale. We already have one confirmed speaker, JB Strauber, co-founder and CEO of Redwood Materials. He's also the previous CTO of Tesla. And we have a few more industry leaders joining us as well. So stay tuned in two weeks. See you there. Bye now.