 A very good morning from Palo Alto here at Stanford University. My name is Wilchu, and together with my great colleague, Itwe, we're delighted to welcome you to another Storage X seminar. Today, we are very excited to have a focused seminar on redox flow batteries. This is one of the key technologies being considered for long duration energy storage and also for electrochemical processing more broadly. Flow battery has been developed over the past few decades. It's a very rich playground for both science and engineering and offered the promise to decouple energy and power when it comes to energy storage and low cost of deployment. This is something that we have been very interested at Stanford, and this is something has received really worldwide attention, especially in the past 10 years. And we are delighted to feature two of the leading researchers in this area, focused not only on the chemistry side, but also the engineering side of redox flow batteries, looking at their fundamental limitations as well as opportunities for technology deployment. And we're very pleased to have Fik Rashad, who is a professor of chemical engineering from MIT, and also Michael Marschak from the University of Colorado, Boulder. And they will do a deep dive today, as I mentioned, everything from chemistry to engineering to system level analysis. So to get started, let me hand things off to Yi, who will introduce Fik. Yi, please go ahead. Well, thank you so much. I would like to add my welcome to everybody to attend Astrogex symposium. So it's a little bit early here in California, I know. It's my great honor to introduce Professor Fik Rashad from MIT. He's currently a professor of chemical engineering at MIT. And he joined in faculty since 2013. So after finishing what he did undergraduate in UPenn and PhD in the University of Illinois at Havana-Champaign, and also did a two-year post-doc at Argonne National Lab. Since joining MIT, his lab has been focusing on advancing the science, engineering, electrochemical technologies for sustainable energy economy. So a lot of exciting research come out of his lab. Today's topic, like Real said, is on redox flow batteries of very exciting areas. Let me welcome Fik to come to the stage to teach us about the latest and greatest fake place to take this away. Thank you so much for the wonderful introduction, Yi and Will. And thank you for the invitation to speak at this seminar. And thank you to all of you, especially those of you on the West Coast who woke up very early to come and see me talk. Today, what I'm hoping to do is to provide a little bit of an overview about redox flow batteries, why they're important for long-duration energy storage, why we think they're a really interesting electrochemical platform for meeting these emerging demands and where we are today and what are the opportunities going forward. So I hope to cover that in the next 30 minutes or so here with you and then open it up at the end for a little bit of Q&A. So maybe to start on a very high level, the availability, abundance and energy density of fossil fuels has really driven progress over the last century and has led to dramatic improvements in quality of life worldwide. This includes the emergence of the middle class in many areas, increasing urbanization and the rapid growth of emerging economies such as India and China. The challenge that we face, however, is that this increasing energy consumption is intrinsically tied to fossil generation. And so we can sort of see net CO2 emissions coming out here aligned with the combustion of fossil fuels in order to drive our modern lifestyle. In order to prevent some of the worst aspects of climate change, we need to start thinking about becoming carbon neutral fairly quickly and ultimately becoming carbon negative in some of these projections even within the next step, within this century. But an even key challenge, bigger challenge here, is that this has to be done on a global scale. It's much larger than almost anything that we've ever had to do beforehand and solutions have to be considered in a way that does not stifle economic development. So we need to try to decouple this energy demand from carbon emissions without reducing quality of life around the world and stifling economic development. So that's quite a large challenge that we face. There has been some improvements over the years. I'm showing here just an example. This is a bit of a Eurocentric example, but it looks as U.S. utility-scale solar power purchase agreements starting from 2008 and going to 2018. This is data from LBNL. And what you can see is that the cost of solar electricity is dropping fairly dramatically over the decade. And you're seeing solar electricity starting off in sun-drenched areas like California and the Southwest. But as we see the cost come down and down, you start to see it pop up in other parts of the country. And if we look at now this projection from the U.S. Energy Information Administration, you can see that we're projecting renewables to increase fairly dramatically over the next 30 years or so, ultimately becoming the primary energy consumption in some of these more aggressive projections here. But still, there's going to be a fairly large amount of petrochemicals, at least for the foreseeable future. This expanding deployment of low-cost renewable resources does offer a pathway to more sustainable electricity generation, but these low-cost masks somewhat of an issue, which is the resource intermittency. So this is data provided by courtesy of Dr. Patrick Doyle when he was at the MIT Energy Initiative, Patrick Brown, excuse me. And this basically shows the hourly variability of wind and solar resources in Texas in 2014. And so on the Y axis here are the months of the year, and on the X axis are each day in that month, where this gap here represents about 24 hours, and this data has been normalized. So you can look at the availability of solar, the availability of wind, and the load all on the same level. And what you'll notice here is that there is intermittency over a range of different timescales, ranging from a couple of hours to a couple of days to even variations in seasonal availability, say in the summer versus in the winter, for things like solar and wind. And this becomes quite challenging in terms of planning out a grid because this intermittency challenges the supply and demand of the electricity grid. And so we have to figure out ways to manage that. Energy storage is one of the ways of managing this intermittency, but other methods include transmission and distribution, infrastructure development, as well as demand side management. All of these should be considered in this approach today. I'm just going to simply talk about energy storage. And so if we wanted to make now renewables the basis of a sustainable energy economy, I think the first thing we need to do, as I mentioned, is firm up electricity generation to allow for baseload power generation that's reliable from these intermitten resources. This may also allow us to advance electrical grid structures in the U.S. and around the world, and even enable microgrids and off-grid type applications based upon the local availability of these resources. If we confirm up the delivery of renewable electricity, this then allows us to begin to significantly decarbonize transportation. While there are battery electric vehicles and fuel cell vehicles available today, at least for the battery electric vehicles, when you plug your Tesla or whatever car you have into the wall, it's drawing power from a thermal generation plant up the road. And so you're simply shifting where the carbon dioxide emissions are coming from. If we can decarbonize electricity generation, that then allows us to take advantage of the advances that exist within the vehicular market with battery electric vehicles. And then finally, if you're able to provide low-cost electrical energy, that offers up a lot of very interesting opportunities within industry using electrochemical processes in tandem or in replacement of thermochemical processes for fuel and chemical production. It may enable new opportunities for process intensification. And it offers opportunity for environmental remediation such as carbon dioxide capture and utilization to generate fuels and chemicals. But all of this, I believe starts with the ability to firm up electricity generation from renewables and the ability to reliably know that electricity can be delivered from these intermittent resources. And so that's why I think long duration energy storage becomes a very important piece of the puzzle in order for us to unlock this sustainable future. So today, pumped hydroelectric storage, it constitutes more than 99% of the deployed energy storage capacity in the world for stationary energy storage. It's a proven robust low-cost technology that's been fielded since I believe the 1930s. And an enabling feature of this technology is the fact that it uses a low-cost working fluid, in this case, water, and then it has turbines that provide power. And so how the system works is that you have water at two different elevations. When you wanna store energy, you take electricity and you put it into pumps and those pumps will push water to a high elevation. So you're storing energy in that potential energy of the water. Once you wanna release that energy and get the electricity out of it, you let the water flow down, it flows through some turbines which spin up and provide electricity to the grid. This allows you to have a relatively low-cost way of storing energy. And ultimately, as these things get bigger and bigger, there's a decoupling of the energy which is the size of the reservoir and the power which is the size of the turbines which allows you to get very favorable scaling. The challenge with these types of systems is the geological requirements. You need to have a large body of water and you need to be able to have that water at different elevations. Also, the relatively large project sizes become an issue even though it might be cheap on a dollars per energy basis, you may have to make a big project. So that's a lot of energy, that's a lot of dollars because it's relatively low energy density, right? And these larger project sizes actually make it quite difficult in terms of thinking about financing and project risk and your ability to get a return on investment. So while these technologies exist, we wanna find other ways to think about storing energy on the grid. And so that's why we turn to electrochemical options. And so when we think about batteries for the grid, many of you are quite familiar with batteries in your everyday life and your portable electronics, potentially in your laptop computers, your cell phones, and in other parts of your life. Most of those batteries are enclosed batteries either lithium ion or lead acid. And the reason we like enclosed batteries is because they're quite energy dense. You're packing a lot of power and energy into a relatively small package. However, for grid scale energy storage, we're willing to give up these size constraints for dramatic improvements in lifetime. We want them to last for a very long time and we want to have a very low price per kilowatt hour. So price per energy. What this means is that we can begin to open up the design playbook going from solely enclosed batteries to things that look a little bit more like hybrid metal flow, which is kind of a mix between a flow and an enclosed battery, where you have one flowing component or a full flow battery, where you have now a very different looking system that looks a little bit more like a power plant than it does an energy storage device. I'm gonna talk primarily about flow batteries here, but I think the message I want to make sure I deliver on this slide is that for grid scale energy storage, different batteries can serve different roles within the grid. For long duration energy storage flow seems to be one of the more attractive options and hopefully I'll get into that in the next few minutes here. So I wanted to start off here with a little bit of flow battery fundamentals just to give people a sense as to how these types of systems work. And so within the flow battery, you have two electrolytes, one positive and one negative. The positive and the negative refer to the redox potentials as for relative to one another. So one is more negative and one is more positive of the other. Within these electrolytes, you have an active species. This is what actually stores the charge through redox reactions. That active species is dissolved in an electrolyte, which contains a supporting salt and a solvent. So you might imagine the supporting electrolyte could be something like sulfuric acid and a solvent could be something like water. And that's used to dissolve this active species and to store and release energy, we are oxidizing and reducing these active species in tandem to shift electrons from one side of the battery to the other side of the battery. Where these reactions occur is within this electrochemical reactor. For those of you who are familiar with fuel cells, this electrochemical reactor has a lot of similarities to fuel cells with an exception that you have liquid redox reactants moving through rather than gaseous reactants. And the fact that you want to charge and discharge rather than operate the system in one direction. This is a blow up of the reactor and I'm gonna focus first on the membrane and then work my way outwards to the carbon paper electrodes and then the flow fields. This is simply one side of the reactor. The other side is the mirror image. I'm not blowing that out to show you as well. The ion exchange membrane that separates the two redox solutions is designed to allow ions to pass back and forth between the two electrolytes to balance the charge that is moving through the external circuit and maintain electroneutrality. But you want this ion exchange membrane to block species crossover, species going from one side of the cell to the other. This species crossover will lead to capacity fade over time in the battery. And that leads to an ultimately a loss of life or an inability to deliver energy at the original marked nameplate capacity. Abutted to this membrane is a porous electrode with a flow field behind it. This is a top-down view of the flow field where the lighter blue channels represent the open channels for fluid to flow and the darker regions represent the ribs or solid blocks. So if you're looking at this blue channel where I'm drawing my laser pointer now, this is an example of this channel right here. You'll notice that this is an interdigitated flow field. You can think about my fingers interlocking with one another. The only way out for this fluid is to go up and over this rib. This actually means that the fluid needs to be forced through this porous electrode by a convective flux. Within this porous electrode is where the electrochemical reactions happen. They're typically based upon carbon papers or carbon felt similar to the gas diffusion electrodes in fuel cells for those folks who are familiar. And within these porous electrodes, you have advective or convective fluxes through these larger pores and then diffusive fluxes through the smaller pores to get to the surface where an electrochemical reaction occurs, an oxidation or reduction, and then the species moves back out of the reactor and back into the tank. This flow battery, this format here enables decoupling of energy and power in the scaling. What that means is that the energy is proportional to the number of tanks you have and the concentration of the reactant within those tanks and it can be scaled by adding more tanks. The power can be scaled by adding more reactors or changing the reactor design, but you can independently specify and modify those during operation. It's relatively simplified manufacturing as compared to lithium ion batteries. Most systems of high durability and low maintenance, I'll get to that a little bit later in this presentation. And finally, there's location independence as compared to something like pumped hydro that requires some geological formations. But there's no free lunch. Redox flow batteries are best suited for energy intensive grid storage applications. What I mean by that is that because we're taking the active species and dissolving it into the electrolyte, we're losing energy density as compared to cells that are more closely packed together, think like lithium ion batteries. And so where the flow battery really shines is that the longer and longer duration you have, this decoupling of power and energy costs becomes increasingly attractive. So the system is ultimately asymptotically approaching the cost of the materials in the tank as the tank gets bigger and bigger. And I'll try to get to that in a later slide. So if you got excited by this technology and decided that you wanted to purchase a flow battery and you looked around, you'd probably find that the vanadium flow battery is the most common option. It's what's most available right now. It's the currency to the art technology. It was advanced in the early 1980s by Professor Maria Skylos-Cazakos in Australia, but there's been a number of wonderful contributions over the years to this technology and a number of companies that have spun out. I'm simply showing an example of a few here, Largo Clean Energy, Invinity, Cell Cube, and Sumitomo, but there are many others that exist in the world today. This is a very interesting technology because it relies on the fact that vanadium has four stable and soluble redox states that can be dissolved in sulfuric acid type electrolytes. And what this allows you to do is have vanadium on either side of the battery where on the lower side of the battery, you're relying on the redox reaction between vanadium two and vanadium three. And at a high potential side of the battery, the positive side, you're relying on redox reactions between vanadium four and vanadium five. The fact that you have this single parent compound actually facilitates low cost electrolyte management strategies. So if this ion selective membrane in the middle is perm selective to protons, but also will allow a certain amount of vanadiums to go through, you'd imagine you'd have capacity fade in your battery. But what the vanadium system allows you to do is simply rebalance that capacity by shifting volumes from one tank into the other to rebalance your system during the lifetime. Also, because we're not alchemists, vanadium is always gonna be vanadium. And so there's a potential for using a vanadium system and at the end of the battery life, recovering that vanadium, recycling it and either putting it into another battery or potentially using that vanadium for other potentially more profitable uses such as integrating it to steel manufacturing if you wish, right? So there is a salvage value for this battery if you wanna take it down, but the vanadium should not decay and that allows you to run the system for a very, very long period of time. However, there are challenges. Vanadium right now has a relatively high capital cost vanadium flow batteries. And that's because of the cost of vanadium right now. Vanadium is primarily used in steel manufacturing to strengthen steel. And because the vanadium flow battery market is not very large, we are price takers for the steel market. There and consequently the system can be quite expensive. There's also significant volatility within the vanadium supply chain right now based upon steel manufacturing only occurring at certain plants and only some plants enable you to get vanadium from the steel manufacturing process that can be used for the flow batteries. So these supply chain constraints may ultimately make it difficult for the vanadium flow battery to reach really, really low costs or for it to be able to reach very large scales of operation. There is work in this area to try to improve this. Some of this work includes things like electrolyte leasing could involve futures contracts or vertical integration between vanadium or redox flow battery companies and vanadium miners. And these are all very good steps forward in terms of reducing the cost and reducing the volatility for the vanadium flow batteries. However, as we begin to look at the promise for redox flow batteries for stationary storage, we can begin to look at why we think flow batteries are really attractive for long duration energy storage and why we think we might need to go beyond vanadium. So this is a graph that I put together some time ago. So some of these numbers might be a little bit off now that lithium ion batteries are improving at a tremendous rate. But what I hope you can take away from it is the fact that lithium ion batteries don't necessarily scale with discharge duration. The way you might think about that is when you want to have longer discharge duration of the battery, you're simply plusing up the number of cells or the number of packs in your system, you're not necessarily having that independent scaling of power and energy that you would be able to do in redox flow batteries, which allows you to go from power dominated costs to energy dominated costs and ultimately asymptotically approach the cost of the materials that's in the tank. But you can see here the challenge if we were have to buy the vanadium upfront in this particular case, the cost of vanadium electrolytes is still going to be not as competitive versus lithium ion batteries. And so this difference in cost may not be enough for you to deploy a vanadium redox flow battery as compared to a lithium ion battery, which is why we're always trying to push towards lower cost chemistries that would allow us to get to these lower battery costs at an earlier stage. And for some of you who might have read the IRA, the Inflation Reduction Act, one thing that was called out is they're looking for batteries with a power to energy ratio of one to 100, which would be right around here that's about four days or so off a discharge. And so you're looking for batteries to be sort of in this area right here. And so much of the efforts in the field has been focused on going away from some of these more expensive transmission salts to thinking about things like organics or metal centered coordination complexes and beginning to step away from maybe acidic aqueous electrolytes which challenges the materials that you can use to construct the battery to think about either non-aqueous electrolytes or electrolytes that have more neutral pH and sort of milder aqueous electrolytes. But hopefully what you can take away for this is flow batteries can have extremely low capital costs for long duration discharge applications provided we can identify these inexpensive redox chemistries for use in these systems. But therein lies the challenge is that the interconnected requirements of the flow battery electrolytes make finding a suitable candidate to replace vanadium quite challenging because you're basically trying to match a system that's gonna have the right redox potentials to give you higher cell voltages. You want fast redox reactions. You want high effective electron concentrations so you can store a lot of energy. That means you're gonna have to have active species which are soluble and stable across multiple different oxidation states. You wanna have designed those systems so they can exist in electrolytes that are conductive and not overly viscous to reduce pumping losses. And those electrolytes need to be compatible with membranes or separators that are conductive and selective so you don't have this crossover. And above all, you need to have low materials costs and you need to be aware of safety. The point is to say that there are lots of trade-offs that exist here in compromises. And the question that we might have is how do we rank order these trade-offs and compromises or which of these properties are going to be most important going forward? And this is where something like techno economic modeling can become helpful. And so what I'm showing here is an example of techno economic modeling where we can think about forward evaluating to look at materials properties, cost parameters and component performance of new things that we might wanna put into a flow battery, throw it into a computer and pop out what it does for system cost and performance to see if we're actually moving forward. However, another way to look at this problem is to solve it in the inverse where you back translate from a desired system cost and performance and you go back and identify what range of materials properties, what range of cost parameters and what range of component performances would allow you to hit a system that had that desired cost and performance. And so a lot of what we try to think about is how do we connect materials properties to system targets, asking are there critical sets of interdependent materials properties that we should be aware of in our design strategies? Asking how do materials design decisions impact other aspects of the flow battery system? So are we solving one problem and creating three more in the process? And are there technical roadblocks that might be obscured at the component level but all of a sudden become an issue when you try to scale up? And so that's how we try to use our economic model to help us understand materials selection. And so just to give you a quick sense of this before I wrap up, if I take a flow battery I can draw a box around it and say what goes into this battery if I wanted to have a price of say $100 per kilowatt hour which was a long-term target established by the DOE some years ago. And so we're looking at price of the battery divided by the energy on discharge of the battery. And what we find is that that should be a summation of the cost associated with the reactor the cost associated with the electrolyte and the cost associated with the balance of plant all the other stuff that you need for the battery to work that's the pumps, the tanks, the heat exchangers all that kind of stuff that doesn't store energy in the system but you need to factor it in if you're going to design the system. And then ultimately we need to consider these additional costs so appreciation labor depreciation excuse me labor overhead a profit margin if you want it to be a price. And so in our modeling we collaborate with industry as well as collaborating with members at national laboratories to really help us out with these balance of plants and additional costs which are difficult to see from an academic vantage point but our main focus tends to be on costs of the electrolyte and costs of the reactor. And we consider present day battery costs for us to understand how to benchmark our model and then we sort of try to push out for future state battery costs and what we're essentially trying to ask ourselves is how low can costs get under conservative to optimistic ranges? And what does that mean in terms of the types of decisions you make in terms of what battery chemistries to pursue in the near term? In a little bit greater detail here we look at the cost of the reactor and we can see that it's directly dependent upon the cost of the materials that go into the reactor it's directly dependent upon the resistance of that reactor and it's inversely proportional to the efficiencies of the reactor operation and to the square of potential. So what that tells you is that cell voltage has a really important role to play in the cost of the reactor. We can also look at the cost of the electrolyte and what that is is the summation of the active materials both the positive and the negative the cost of the salt and the cost of the solvent and all of this is also inversely proportional to the potential and the efficiencies of the system, right? If I have an aqueous system say like sulfuric acid and water it's harder to get cheaper than something like that. So many costs in these systems are described by the active materials. But if I have a non-aqueous system that might give me higher voltage the costs of the salt and the solvent are non-trivial and need to be considered. And so I just wanna give you a little example here before I wrap up about how one might begin to use these types of techno economic models to help to guide design. And so the model allows me to plot any number of variables as a function of the other but I have a hard time understanding more than three variables at once on a plot. So what I'm gonna plot here today is cell voltage on the y-axis area specific resistance on the x-axis remember cell voltage is in the power cost and the energy costs and the area specific resistance is directly proportional to the cost of the reactor. And I'm going to also plot contours of isomolality that's moles per kilogram to identify design spaces that are favorable for non-aqueous redox flow batteries first and then I'm gonna show you aqueous flow batteries next year in a second. And in the case of the aqueous flow battery you'll see that I'm looking at a typical stability limit of about 1.5 volts which is about the kinetic stability of water under acidic conditions on a carbon based electrode. But what I think we'll see in the next presentation is that that number can be pushed a lot higher through some very clever chemistry and engineering approaches. But I hope you take from a graph like this however is that they're clearly trade-offs within this design space if I get to higher voltage type systems I'm able to get to lower concentrations of redox active species let's say in the non-aqueous design space and if I get to reactors that are too resistive it becomes harder and harder for me to meet my design goals because the cost of a reactor becomes too high. We can then extract now some key metrics from plots such as this and so we can start to say well we wanna have active species costs for flow batteries that are gonna be relatively low in terms of dollars per kilogram. We also wanna make sure that the molecular weights are relatively low as well in terms of grams per mole and this is because we wanna have relatively low kilograms per mole of electrons stored. If we have molecules that are too big even though they're cheap it's gonna be still a lot of kilograms in order to get enough energy storage. For aqueous systems we want high cell voltage and low ASRs this will get us into this sweet spot right here where we don't have to have really, really high concentrations and for non-aqueous systems we need to have relatively high active species concentrations relatively high cell voltages, low resistances and a low salt cost factor because largely salts under non-aqueous conditions are fluorinated and fluorination is an expensive process. So when you look at plots like this this is very helpful because it gives you some set of targets you can go after but it also gives you a census to risk. And so when I look at a plot like this as an academic I'm very excited about this non-aqueous space because look at the size of design space look at the potential opportunities we can have to design low cost batteries that can meet this target through a range of different conditions. However, you can also look and say well are there any non-aqueous flow batteries that are commercialized right now? The answer is I believe no. And so you say there's a lot of risk here there's a lot of problems that we need to overcome in order to access this larger design space and so this might take a longer time for us to begin to hit this target. However, there are aqueous flow batteries that exist right now I just talked to you about a vanadium redux flow battery. And so maybe those are systems that are gonna come first this is gonna be the next generation of flow batteries where you replace vanadium with some interesting lower cost chemistries that we can design towards. And so if I'm maybe an industrially focused person what I would say is that there's less technology risk in the aqueous redux flow batteries and therefore maybe that's the place that we should look to first and then in the longer term look at non-aqueous redux flow batteries. And so then you say, well, okay, that's great. What are we looking for? What should we go after? And there are two potentially different pathways to get to inexpensive charge storage materials. The first is these commodity scale inorganic materials so think about things like sulfur, iron things of that nature. And the benefits that you get from these types of systems is that they're earth abundant materials. Many of them already have a materials production infrastructure or they can be recycled or reuse of waste products from that structure. That's an example of this sulfur over here coming from the tar sands in Alberta, Canada. Are we able to take that sulfur and use it? This is also one of the reasons why people are so interested in lithium sulfur type batteries. The challenge with an approach like this, however, are things like upgrading requirements. How do you go from a sulfur pile like this to something that might be an electrochemical grade sulfur? And what is the cost associated with removing those impurities and how important is it to get all of those impurities out? Also, some of the technical attributes of the system might be difficult to work around. For example, the redox reactions may have relatively low efficiency. There might be, say, kinetically sluggish. There might be lots of side reactions you're worried about and the reversibility might become an issue. And so you need to think about designing systems. It could be new catalysts. It could be new operating envelopes. However, all of that then can lead to cost of these other system components. And so you might have a really cheap base component but it might ultimately become a very expensive system if we don't think about the design very carefully. Another option is engineered molecules via organic chemistry. And so it could be taking naked organic molecules or taking those molecules and complexing them around maybe some cheap metal-centered coordination complexes. And so this could be another approach. So it's organics or metal-centered coordination complexes or options. The benefits here are you can have tuneable technical properties based upon tweaks of organic chemistry. I can change things like the solubility, the redox potential, potentially the stability. You can have it based upon abundant constituent elements and there's also a potential for mass production if we're clever about the organics that we use. Many organics are synthesized at a very large scale either from petrochemical products or through other product development pathways. And we can also identify potentially cheap cores or say metal centers that could be used for the complexes drawing again from these commodity scale species. The challenges that this area faces however is ultimately molecular stability. These materials can decay and they can decay in relatively complex ways that are difficult to predict a priori and difficult to manage during operation. And the potential cost and scalability of a desired high-performing targets can become an issue. And the issue might be that to get an organic that's gonna be stable enough I need to put a lot of decorations on it. And the addition of these decorations adds cost because it's multiple synthetic steps and it also adds weight to the molecule. So it becomes less and less charged dense and that then leads to buying more kilograms in order to meet a particular capacity target. So these are areas that we would need to work on. And so I'm towards the top of my time. And so I just wanted to finish up here with a couple of pathways forward just hopefully to excite maybe some discussion after this. The first is how do we accelerate discovery of new organic molecules or new redox active components for our battery system? And this was something proposed by Alan Asparaguzek in his group a number of years ago, about 2018 where they pointed out that the current paradigm for discovery is a materials concept, molecular synthesis, a device construction and then testing and then looping back around this feedback cycle ultimately to get to scaling and manufacturing. And they postulated that there might be faster ways to begin to do this through inverse design, using generative processes, simulation and optimization and then an integrated pipeline that took advantage of software and robotics advances to more quickly test materials, extract valuable information and then loop back around hopefully reducing the cost and time for this scaling approach. Also within Jay Caesar, which I'm a part of people are also thinking about this this is a graph from Professor Jeff Moore who leads the Reduxmore effort at Jay Caesar, thinking about a little bit about seeding new molecules using chemical intuition and then entering them into this design loop where you can use ML or machine learning to expand the design space, automate the synthesis, perform high throughput characterization, assess the suitability to multi-objective targets and then dump it into databases which are supported by high performance computing to allow us to accelerate this discovery loop. And I think advances in these areas can hopefully allow us to get to better molecules more quickly. I think stepping away also from bare transition metals that are dissolved in electrolytes offers new design strategies going from things that are Redux active monomeric units to oligomers, polymers and colloids. What you buy when you go to these larger molecular structures is you get to move away from ion exchange membranes that might be expensive or might and might also not be completely perm selective. And so some larger molecules allow you to go to things that look a little bit more like nanofiltration membranes and ultimately up to things that look a little bit more like coffee filters. And what this may allow you to do is to get to cheaper overall battery designs this way. The challenge, of course, is as you go up this design scale, you're gonna have issues with transport, Redux, kinetics, solubility and cost. And what we need to ask ourselves is, is there a sweet spot along this axis where we wanna design? Maybe it's still down here with the molecules or maybe it's a little bit further up. These are open areas for analysis. And then finally, if you want a long duration energy storage system to work you need it to last for a very long time in order for you to recoup your investment. It's oftentimes operating lifetimes of 10 to 20 years. And so you might ask yourself, well, how do I know the battery is gonna run for that period of time? And I'm gonna be able to recoup my investment over that long project lifetime. And this is simulated data showing just some decays that can happen within Redux flow batteries. And those decays can be complex because it could be due to a number of different reasons that you need to piece out. And so what we're trying to figure out here is what are modes of performance degradation in batteries? And how can we predict the lifetime of these systems without having to operate the battery for that lifetime? And so I wanna wrap up here and hopefully what you've taken away from the past 30 to 35 minutes is that cost-effective energy storage can facilitate a transition to a sustainable global energy economy. I think long duration energy storage is gonna be a very important part of this. Flow batteries have shown promise for these applications provided we can identify inexpensive electrolyte chemicals and provided balance of plant designs can be identified that are also inexpensive. I think early stage techno-economic modeling does offer an opportunity for us to inform research directions and highlight technical bottlenecks at an early stage allowing us to move perhaps faster to these new systems and to quickly go through the design space in a much more organized fashion. But techno-economic modeling alone can't get us there. We also need advancements in the computational tools that we use and the experimental platforms that we set forth which are necessary over multiple length and time scales in order to advance these new technologies. And with that, I thank you for your attention. I'm gonna briefly acknowledge my research group, my funding, a group of wonderful collaborators in industry, national laboratories and also in the academia. And also thank you for your attention. And I hopefully I've left enough time for a few questions before we get to the next talk. Thank you so much. Well, thank you, Fik. This is a very interesting talk. Thank you for sharing with us about these multi-parameters and also techno-economic modeling, the importance of that. So while you are right on time, let me ask you a few questions that later we'll go into together with the second speaker, Michael, on the panel. You give a really nice overview on how to think about Redox Flow batteries. We know this has been around for decades. So it's good to analyze what are the key challenges based on the past learning. Let's use a Vanadium Redox Flow as an example. If you will pick, or first of all, update everybody or the audience, what's the current cost of the Redox Flow for Vanadium at this moment? Maybe we calibrate ourselves a little bit first, Fik. So that's a really great question, Yi. And before I answer it, I'm going to provide a caveat, which is there are many different small Vanadium Redox Flow battery companies. And so a lot of times it's difficult for an academic like me to get a sense of the cost because many of them are competing against one another to sell their product. And so you're not necessarily going to get the exact cost information from them. That said, the numbers that I've seen or the numbers that have been reported, and for those folks who are interested, you can go to things like the Lazard report that gives a good sense as to where the battery costs are right now. In terms of capital costs, it's probably around, I would say, maybe $300 to $400 per kilowatt hour right now. I'm happy to be wrong about that. Remember, I'm an academic, not somebody who's trying to sell these kind of things, but this is around where I've heard the costs are today. Okay. So I'm looking at the baseline of this cost, right? So what are the major components that's the most costly? Then you say, well, how I'm going to improve it later, right? So what are the major components right now? So that's just still vanadium as an example. Yes, I think the major component is the cost of vanadium. And I think that the challenge is that vanadium is not only expensive, right? To get the vanadium electrolyte, but also the price of vanadium is fairly volatile, right? Which means that the cost can vary depending upon time quite a bit. And this means that oftentimes it's difficult to sort of price the battery, especially for a small company that might be assembling the cell, buying the vanadium and essentially mixing up their own electrolyte locally, right? So that's one of the major challenges is essentially sort of the risk associated with the cost of vanadium and being able to get the financing. I would say the other challenge is within the reactor itself. The way you might think about it is that everybody sort of all these smaller companies have their own special sauce of building the reactor, right? So everyone's is a little bit different, which means that a lot of these designs are sort of very similar on the very high level, but kind of boutique on the local level. But almost all of them, I believe all of them will have a napheon membrane in the middle, which is a chlorinated membrane that has largely been used in PEM fuel cells. Napheon is relatively expensive right now. It's produced at relatively low volumes, and it actually becomes a relatively large cost component in these systems up front as well. So the cost of the napheon membrane, I believe, is the largest cost component within the reactor in most of these systems. Yeah, I think I really like analysis on the isolated membrane. So vanadium system, one of the very, I think the one's company is Rong Ke in China. They are having this big factory right there. I remember the professor Zhang right there mentioning the membrane innovation they have moving away from napheon in order to reduce the cost. Fik, did you look into that? Did you have information about that to see where the membrane innovation can help reducing the cost? I don't have very much information on Rong Ke power. I know that Rong Ke power was the one of the larger companies in this area. And I remember I was fortunate enough to go to a conference in China and visit one of their showrooms. And the scale in which they're operating and planning operating, this was a number of years ago, pre-COVID, was absolutely astounding. And what you can imagine is that some of the cost come down in these systems. So kind of bringing the cost down to the battery has to do with economies of scale. And so if you can make enough batteries and you can make them efficiently enough, the cost can come down fairly significantly. I don't know too much about their membrane innovations. I think that there is a lot of room in that space. The way I would think about it would be, the vanadium system, especially vanadium five is a fairly aggressive material that can begin to attack membranes that are not fluorinated. If you can figure out a way to prevent that vanadium five attack on the membrane and the membrane degradation as a result of it, you might be on to something. And the reason I say that is because you're not as worried about crossover in these systems because you can simply rebalance the cell and it then becomes an operational and economic argument. How often do you want to rebalance and what is say the lost opportunity to make money on the battery? But that can be engineered out, right? So I believe that's a good pathway forward. If they have some advancements that people know about in the audience, certainly please feel free to email me. I'm all ears. That's good thing. So I'll ask a couple more questions then we can move on to Mike's talk. So your analysis is multi-primary is just fantastic about, you know, this cell wall teeth, this solubility, like molecular weight and every specific resistance. So throughout this analysis, you have a system you mentioned, for example, sulfur, iron, this also organic molecules. Any one or two or few particular redox system now stand now to say, well, this is quite promising. I think there's a number of choices, right? All this will be interesting to know, well, what's fixed analysis? What's the answers coming out of that? Yeah, are there your favorite molecules right now? Well, I think we're about to hear about a really promising coming up here. So I'm not going to steal Professor Marshak's thunder. But I will say that some of the more promising chemistry that have come out are sun catalytics which was bought out by Lockheed Martin a number of years ago has been advancing a metal centered coordination complex chemistry that looks quite promising. I think it's going to be similar broadly speaking to what Professor Marshak is going to talk about. And I think there's a lot of design innovation there. That operates under aqueous conditions. I also think there's a tremendous amount of opportunities in the organic space for things like quinones. So thinking about the work from Professor Mike Aziz and Roy Gordon as well as others who've really been to push the limit in terms of the solubility, stability and longevity of these quinone type species. The challenge on that side is that it's hard to find organic molecules to pair it with on the positive side. Quinones work really well at lower potentials but for higher potentials largely we're still using coordination complexes, things like ferrocyanide or other molecules like that. In the non-aqueous space, we're still on a good for a bit back. The TRO level is very low in that area right now but there is a lot of really interesting innovations coming from JCZR, which I'm a part of. So maybe I'm a little bit biased there but also other areas around the world. Really I think pushing the envelope on molecular design but I think once we can come up with archetypal chemistries in that area and I think we have a few right now, then we can really begin to push the cell engineering to see how good these systems can become. So I'm excited about that area too but think that it'll be coming after some of these aqueous innovations really start to take off. Yeah, sounds cool, maybe one last question then we can move on to Mike. So you also mentioned the suspension, the redox with particles, certainly very early days, yaming chance, 24 am. And it's an idea really, really interesting, certainly has its own challenge. Well, this can certainly simplified as like the main brain, like you said in your presentation you can go to the pore size that could be bigger. So you save in the main brain and also you potentially can utilize very different redox is a solid. So any analysis on the suspension redox direction, the catholic and the like, yeah. So not at this juncture we're working on it though but what I can say is that suspensions are really interesting for a couple of different reasons, right? So you had mentioned 24 M for those folks on the line who are unfamiliar, the concept originally here was that you could take lithium ion type intercalation materials and suspend them in an electrolyte with a carbon fractal network that would then allow you to charge and discharge and have a flowable deformable electrode that could be pumped through a reactor. What ultimately becomes a challenge in a system like that is that it's a non-Newtonian fluid. I think it was actually a Herschel-Buckley fluid which means that it has a yield stress and then it's shear thinning which actually makes it very difficult to design reactors for something like that and not have your pumping costs go through the roof. Another area that you could use and this is work pioneered by Qing Wang I hope I'm pronouncing his name correctly at NUS has been doing a lot of very interesting work on what's called redox targeting where you have those energy storage materials that are stacked in the tank. And so you have redox materials almost as redox mediators that will pick up an electron from the reactor, run it over to the tank and dump it off on a solid state material. Those are both very interesting designs. The techno economics on those are still out right now but I hope to be able to come back maybe in a couple of years and let you guys know what the space looks like over there. Yeah, well, I agree. I think Qing Wang's approach is really interesting right there, you know, that's actually very smart design. So think with that, Wang, thank you for your presentation and Q&A will be back in the panel discussion. Will, I'll back to you. All right, thank you very much, Ian and Fik for the wonderful presentation and introduction to the rich playground of redox flow batteries. I'm delighted to introduce our second speaker, Michael, if I can have you join us at the stage, please. Good morning, Michael. It is my great pleasure to introduce that Professor Michael Maschak from the University of Colorado Boulder. And he has been working in the area of redox flow batteries for more than a decade and has some really exciting results to share with us on new chemistry that can be used to further to technology. In addition to being a professor at UC Boulder, I wanna highlight Michael's contribution to science in two different areas. Upon a learning more about Michael's background, I discovered this wonderful article in the journal Science, not about science by itself, but about the journey to science. And I encourage everyone to read it. Michael wrote a very nice piece on his journey to where he is today. The second thing I wanna highlight is that in addition to being an academic, Michael is also a practicing entrepreneur trying to develop and commercialize the technology. And he is the founder and CTO of Vatoro, which is commercializing the technology that he will be discussing today. Michael, we're really excited to hear about the new chemistry. The floor is yours. Well, thank you Will for that excellent introduction and thank you for the invitation to speak here along with you. And I'm really excited to share some of the work that I'm doing here. I also wanna thank Vic for providing such an excellent introduction to energy storage and flow batteries more specifically. And also just calling out some of the main challenges that when I started my independent career here at the University of Colorado, I sought to address. And so with that, I wanted to start without too much review but just to point out that there is this long duration storage shop at the Department of Energy. This is an earth shot. It was recently written that to date there does not appear to be a candidate for storage technology that simultaneously exhibits every desirable characteristic including high efficiency, high cycle life, low cost and geographic portability. So what I think this really highlights at a high level is that the technology is really not there yet. And there's a lot of work to be done to develop various energy storage chemistries and technologies to get us to get to this point that it's not just a deployment problem. So I'm not gonna speak too much about the idea behind flow batteries. I think Vic did a really excellent job describing this. I would just point out on the right side of this is a picture of a flow battery in our lab. So you can see there's an anode, a cathode and some tubing that pumps the analyte and the liquid analyte and liquid catholite in and out of the cell. And I know that I'm speaking to a community that's very familiar with lithium batteries. And so I wanted to start by addressing some of the major questions I get from people in the lithium field. I put this goops all electrolyte in here. Essentially all of the liquids store all of our energy. There are no solid phase changes. There's no mechanical degradation. You very rarely see as Vic described, capacity loss. Vanadium ions are not gonna undergo an alchemy type reaction to go convert to something else that these can be recycled. That the electrodes themselves are typically just simple carbon electrodes. There's no platinum. There's no other precious metal catalyst typically on these very often the flow batteries are aqueous. So they're intrinsically non-flammable. And as many people understand, water-based chemistries have a very high heat capacity. And so there's inherent thermal control. These batteries can't overheat too much because you're constantly flowing water into them and out. So I think that that's one piece of this that I think is really critical and important to understand before I sort of launch into this. The other thing is energy density. I think there, I constantly get questions. Well, you know, with energy density, they're never gonna compete with lithium because they're not as dense. Well, it turns out first off that energy density is really not a primary cost driver for stationary storage. The second is that comparing an actual lithium cell energy density with the volume metric energy density of a flow battery is really not meaningful. And the reason there is that you have to start to consider in a grid scale storage, the footprint of the system. And if you look at grid scale lithium battery systems, like for example, this photo of a system in Australia, what you'll see is that these are not piles of lithium ion battery cells stacked in the way that you would find packed in such a dense way like a cell phone. These are spaced out in part because there needs to be things to address thermal management. There are power conversion systems. And also each one of these containers needs to be separated by a certain distance because they wanna prevent runaway and when one catches fire, they wanna prevent one container from catching another and starting a chain reaction. And so aqueous flow batteries, water-based flow batteries because they're non-flammable can be stacked much more vertically. If you think of a large tank as I'll show next, because you can have a vertical component to this, you can have actually a much higher system energy density. So if you're looking at what would be the footprint for a 500 megawatt hour system, a flow battery is very competitive and in fact can be lithium for that type of footprint even though the watt hours per gram or these other energy density metrics are not necessarily in any way comparable, flow batteries will always be lower energy density because by definition, we have a liquid in there. We have all electrolytes. So this is never gonna be competitive for transportation, for example. So getting into some of the details here, this is a picture here of an inner vault system. We're working on commercializing iron and chromium flow batteries. Iron and chromium is something that I'll be talking about mostly today. However, in this case, for this chemistry, this is just the simple metal ions dissolved in concentrated hydrochloric acid. And there's problems with that. First off, it's a relatively low cell potential, 1.2 volts. And as Fick said earlier, we really wanna get that voltage up to 1.4, 1.5 or even higher if we can. Next, chromium ions tend to generate hydrogen. And so that results in this relatively low efficiency. The redox kinetics are slow, which resulted in power. And finally, you've got things like concentrated hydrochloric acid, which is volatile and very corrosive. And so that creates issues with additional costs and sealing, gas getting, et cetera. So again, the most common commercialized vanadium flow battery system is based on this vanadium here. It uses vanadium and the property that exists in four different oxidation states. So you can have the same metal ion species on both sides and not worry about crossover. So in thinking about what are the significant barriers to the adoption of flow batteries moving forward, I turn to this academic economic analysis. It's a bit dated from 2015, but I think it demonstrates the points really clearly. On the left, you see the cost breakdown of vanadium flow batteries, where the green wedge is the cost of the vanadium electrolyte. On the right, you see the cost of the iron chrome system where the green is the cost of the iron chrome electrolyte, much smaller. However, because of that lower performance and lower efficiency, all of the other cell components, including the membrane, the plate, the electrodes, as well as some of the balance of system costs end up becoming more expensive in comparison. So what we really need to solve this problem is new materials that can provide both high performance and low cost. So I think that's the key high performance and low cost. And so in looking at this myself, I started looking at the same problem of getting to high voltage. If you can get to high voltage, oftentimes it's talked about that you can get higher energy density. However, I've already mentioned that energy density is not a primary cost driver here. And there's only a one-to-one scaling factor. If you double the voltage, you only double the energy density at the same concentration. So that's not quite so valuable. As Vic mentioned, the power is really where you start to see some benefits. And that's because of this relationship that the power in watts per square centimeter is related to the square of the voltage over the resistance. And so here now you start to get this power level of relationship. And finally, if you start to look at the efficiency of the system, another important cost factor, the open circuit voltage or the equilibrium self-potential of the chemistry is highly related or correlated with the energy efficiency. And so here, for example, is a plot where you can look at the open circuit voltage on the x-axis versus the relative round trip efficiency, DC to DC, and that you can now cycle at higher power densities and higher current densities at a higher open circuit voltage with higher efficiency. So I think what this highlights here at a big picture is that getting to high voltage buys you a lot of capabilities downstream. The other thing I wanted to point out here is that there was this myth that's come up over the last, I would say several years, maybe a decade, that aqueous batteries can't support high voltage. And I'm not really sure how this has pervaded the academic literature, considering most people will start their car with a lead acid battery that operates at an open circuit voltage of 2.1 volts per cell. But I also wanted to point out here that in terms of water, if you were to try to break down the thermodynamics of water, one electron at a time, it turns out that the theoretical water splitting window to generate not hydrogen and oxygen, which are multi-electron processes, but the one electron reduced hydrogen atom or the one electron oxidized hydroxyl radical, you end up with this very high theoretical window of five volts. And so there's really, if you can maintain a situation where you're only transferring one electron, you can, in theory, gets a very, very high voltages in water. And really there is no advantage to a non-aqueous electrolyte in terms of the stability window, I would argue. The other thing I would point out is that when we talk about things like stability, I think people are often confusing kinetics and thermodynamic terms. Stability often is going to be, by definition, a thermodynamic term. And if we're just comparing thermodynamics, there's nothing more stable than water that's a liquid. And that may seem like an overstatement, but it's not. I mean, if you are digging things out of the ground, you're finding water and maybe some hydrocarbons that are on their way to decomposing to non-liquid forms. If you take water and heat it up to 10,000 degrees into a plasma where all the oxygen and hydrogen atoms are scrambling around and then you cool back down, you're left with water. There's nothing more stable than water. Whereas if you did that with any other organic solvent, you're gonna end up with a lot of other products that are decomposition. So in terms of the stability, I think, water is the way to go in these. And I'm happy to enter into some kind of debate about that, but anyway, this is just to argue this point a little bit further. If you take a fornated diamond, boron-doped diamond electrode, it was demonstrated almost 20 years ago that you can see a five-bolt window. And so if you can basically shut down all of catalysis, you can observe these very, very large water windows. And so this sort of inspired me, how could we shut down water splitting? How can we, instead of trying to split water to make hydrogen, which is a big thing right now, how can we stop it? How can we design molecules from the ground up that prevent water splitting? And I was inspired by this idea of key leading agents. And so this is a table that describes various metal ions on the left side here, their redox potentials, and then their binding constants to EDTA. So EDTA is a common key leading agent. It's found in a lot of different consumer products, industrial products, and it has this property that it has a very, very high binding constants to various metal ions. So what this means here, 25.9, means that for every 10 to the 25.9 atoms of vanadium, only one will be not coordinated to EDTA in a solution where EDTA and vanadium are in a one to one ratio. And so these very, very high binding constants drive molecular stability. And they also result in this shift from a vanadium redox couple, vanadium 3, 2 of minus 0.6 down to minus 1.03. So these shifts in redox potential resulting from the coordination of this organic key leading agent can result in these dramatic shifts in tuning the redox couples that we see of these metal ions. So the other thing that sort of inspired me was looking at what limiting the choice of atoms to ones that were already produced globally at scale. And if you look at this plot, the y-axis annual production in log of kilograms per year. So what you see here is that you can see the vanadium is quite low. It's on scale with cobalt. And so for those of you in the lithium community who understand already some of the challenges with scaling cobalt, vanadium is actually a little bit worse than that. And so if you're using vanadium not only for your cathode but your anode, then you're running into some very fundamental issues and limitations with the scalability of that element in addition to cost and other things that Vic discussed earlier. So what I was interested in is limiting to things like sodium, potassium, iron, chromium, these very low cost materials. Turns out chromium is second only to iron in terms of the transition metal elements that in terms of annual production with chromium just slightly exceeding copper and then also manganese and zinc. So chromium is the number two second iron in terms of annual production. At least it was in this 2012 report. So the other thing that was sort of the insight that I got here was looking as a chemist at how does EDTA coordinate and recognizing that on the left here is a chromium atom coordinated to EDTA. And look, there's this arm that's hanging off and that's allowing this water to coordinate. And what the challenge there is, is that as soon as water coordinates, you create a pathway for catalyzing water splitting and hydrogen evolution. And so what happens is that as soon as you allow the solvent to touch the metal where the electron is being stored, you can then create catalytic pathways for water splitting. This is the same way that if you think about a lithium ion battery, you don't want the electrolyte, often these polycarbonate or carbonate type electrolytes to actually be in contact with the electrode. You wanna have some SEI type layer, some barrier to protect that electrode from interaction with the solvent. So for the same reason here, this was a problem with EDTA. And so what we did was we moved to this other one that had just a single extra carbon. So in here, these blue atoms are nitrogen. There's two carbons between them here. These blue atoms are again nitrogen, but now there's three carbons between them. This one we called PDTA. We can purchase it in large quantities. It's manufactured at scale. And this solved the problem. This allowed for the right coordination and binding to the chromium ion. And that results in much more stable complex and much more favorable characteristics for flow batteries. So again, we can purchase this, we can make it. The synthesis is very simple. And part of the reason it's so simple is that these keelating agents bind the metals so strongly that you mix them together and it's only gonna make one thing. It's gonna coordinate to the metal ion very, very well. So we can get to relatively high concentrations. And I would also point out what we do get because it's chromium. Chrome, the name comes is derived from the fact that chromium often exists in very bright colors. So we get these really nice color changes similar to the vanadium system as well. And so sort of our initial report here in 2019 showed the difference in a single cycling between the PDTA and black. That's the one that the sort of our champion along with EDTA and red. And what you see here is that for the same charging cycle, EDTA doesn't provide as much discharge. The reason is because EDTA is in constant decomposition generating hydrogen. And in fact, it's generating so much hydrogen in this case that the pH starts to shift so much that the entire cell shuts down. So we can't really do more than a couple of cycles anyway. And so the PDTA really seemed to solve this problem. We got quantitative, coulombic efficiency and very high voltage efficiency. So looking again at our sort of initial report here these are sort of linear sweep of voltammetry that shows the charge and discharge of our battery. You'll see these waves oscillating as a result of our pumps operating. And this was sort of the first demonstration that we could cycle and test a battery that was up at 1.62 volts. We got out very, very high power in this case. This is the very first iteration of this and I'm gonna get to our latest cell performance in a bit. And I wanted to point out here we're using this common catholite liquid containing iron and cyanide, ferrocyanide. Cyanide is often considered to be very toxic but it's toxic because it binds so strongly to iron in your body. If it's already coordinated to iron in this molecule it's actually safe as a food additive. And you've probably consumed and eaten ferrocyanide in some form over the course of your life. So we were able to get out quite stable cycling. In these cases I didn't show the capacity which is often what people wanna see in the lithium community, there was no capacity loss. We were never observed any. I've now started to add that back in to more updated plots because people do wanna make sure that we're not kind of hiding something somewhere. But what I wanted to point out here is that this plot is from a review in 2011 showing kind of a whole series of various redox couples. You can see the Vanadium 3.2, the Vanadium 4.5 here. Our work kind of shattered this hydrogen energy window by sort of complexing the metal line and preventing it from splitting water at this molecular level. And again, the chromium redox couple normally here has been shifted down as a result of this stable complexation process. So in looking at this, I thought, well, how can we just go for a record? And I looked at bromine and I know that bromine is toxic, it's corrosive, it's volatile, it's not really fun to work with. There are batteries, they're bromine. But I thought, well, let's just go for it and try to get to a two plus volt battery just to see what we can do in water. Here is a single discharge, relatively slow, current density of 10 milliamps per square centimeter that's relatively slow for a flow battery. However, we were able to get nearly the full discharge at over two volts, which was really, really exciting. We were able to cycle this at faster 100 milliamps per square centimeter and it's cycled quite reproducibly and stably to high states of charge and many cycles. However, again, bromine is not necessarily something that I or my students enjoy working with in the lab. So we kind of turn back towards the ferrocyanide because it's relatively non-toxic. So this is again, the bromine data, but the point here is that we can get to the highest aqueous redox flow battery voltages over two volts just using this kind of chelation approach. Another thing that we've kind of done in our group is, and I think this highlights some of the interesting facets of flow batteries is we've been able to introduce a flow through cuvette sort of in line with the inlet and outlet of our batteries. And now we can monitor those color changes from magenta to blue actually quantitatively and we can use that to monitor our battery state of charge and some of the health of the system. So it turns out there, there's a couple of points here where the both oxidized chromium three as well as the reduced and fully charged chromium two have the same molar absorptivity. And what that means is that if we monitor that point we can monitor the total concentration of all of our active species as a function of cycling. And what you see here is that we get this oscillation in blue, this should be flat, however it's not. And the reason for that is because as we charge and discharge what we're seeing is water being transported across the membrane along with our ions our sodium and potassium ions that we're using. That said that we do have a net decrease that looks like degradation, however it's not what it actually is, we're not losing capacity here we're decreasing the concentration because there's a net migration of water for one side of the other. And so we use this technique to try to balance the water activity on both sides. So the amount of water on both sides will eventually come to some equilibrium and we wanna be able to control that. So this is an interesting approach to doing that. We can also monitor pH in real time we can see as we cycle our battery that we get these oscillations in pH even in a buffered system. And we're trying to get to the bottom of understanding exactly how and why this is happening. We think there may be actual buffer molecules moving through our membranes or that the membranes may be more conductive to things like H plus rather than potassium plus or sodium plus. So there may be some interesting things that we're currently working on understanding this. We actually have structurally characterized these are single crystal X-ray structures of both our oxidized and reduced chromium ions. And the kind of nice thing to see here is that they look identical except for the fact that one has a calcium and one has a potassium ion. So you can see this is overall a two minus charge and this is an overall one minus charge. They are different. But the fact that they have very little changes in bomb length is why we can get out very high power here. There's very little reorganization. The atoms don't have to move in terms of going from one redox state to the other. And so we can put an electron in and out very, very rapidly resulting in high power without having to rearrange molecules or atoms in very complex ways. We have done a ton of solubility studies looking at various cations. We have screened these as well, not only for solubility but for ionic conductivity, viscosity. And that's another sort of kind of optimization that we've been working on. And so that brings me to sort of the, the moving towards the end of the talk here and talking about sort of some of the bigger picture pieces. We're operating at a neutral pH seven now. That's non-corrosive and we can get away and with non-fluorinated hydrocarbons both in the membranes as well as the gasketing. So using simple hydrocarbons both for the rubber gaskets as well as the membranes not only reduces the cost but I think one of the really important things that's kind of happening in a lot these days is that fluorinated materials are getting more and more flack for contributing these forever chemicals in the environment. I think it's gonna get harder and harder to manufacture these in a safe environmentally friendly way. And so being able to get away from those types of materials not only provides a massive cost savings of the overall system but I think it also provides a much more environmentally friendly and safe approach to this in the long term. Next, I mean in terms of the non-toxicity as I mentioned ferrocyanide can also sometimes be called the yellow prostate of soda. So if you've ever had a salt packet here is an example you'll see yellow prostate of soda the reason that they call it this is because people seeing ferrocyanide might get a little bit concerned. EDTA is also a food additive. Here is a jar of mayonnaise in my kitchen and you can see here one of the ingredients calcium disodium EDTA. So these are also used various key lighting agents are used in things like chelation therapy. So these are actively injected into people for beneficial uses. The other thing I would point out is you might say well what happens when you put chromium in there? Well chromium in this radioactive chromium 51 form is actually used to measure kidney function. So it's the GFR the filtration rate of your kidney. So this material is actively injected into humans safely. I'm not sure you can say that about too many battery materials let alone chemicals overall. And finally EDTA there's a lot of every time I mentioned chromium people think of chromium six Herobrackovich, et cetera. EDTA is actually used environmentally to remove chromium six from the environment. So EDTA will help catalyze and support the reduction of chromium six to the benign oxidation seeds of chromium three. So the EDTA can actually prevent the formation of these really toxic forms of chromium. Finally we're non flammable we're water based and we've essentially cut down hydrogen evolution at the molecular level. We've bound up the metal ions and we've prevented the water from getting in there and reacting. We had an RBE project. I think a group was part of this as well but we were specifically collaborating in this case with Raytheon Technologies, RTRC research center and they tested our electrolyte over months nearly 700 cycles. They were heating the cell they were not heating the cell they were turning on a recombination cell they were turning it off. Trying to break it, it didn't break they got no loss in capacity. You can see here there's no net loss in performance. Can't observe material degradation. One reason for that I think is that EDTA is actually considered by the EPA to be a persistent organic chemical in the environment. Probably not great for the environment but for battery it's great. It's actually known and been studied to decompose only by exposure to UV. So obviously in a battery setting it's very easy to prevent UV degradation of your battery materials. So here's as promised the similar latest performance data. Again, we see zero material degradation. You can see here's just a whole bunch of straight lines. These are various cycling parameters along with efficiency on the left we can get out over 80 almost 85% at 300 milliamps per square centimeter almost 90% round trip efficiency at 200 milliamps per square centimeter. We can get out these power curves that look like the fuel cell community produces and hit at high states of charge 1.63 watts per square centimeter. These are sort of on par with some of the top fuel cell type performance. And we can hit that because we're hitting these very, very high voltages and we don't need to rely on catalysts for our electric chemistry. We're operating again at pH seven. We've now gotten the concentration in a cycle and demonstrated sell up over 1.6 molar results in over 40 amp hours per liter. We can maintain very, very high current efficiencies. Here are a couple of cycles here where we're now cycling at 200, 300, 350 and even 400 milliamps per square centimeter very, very high current density. And again, very good current efficiency, voltage efficiency and round trip energy efficiency. And again, here's this diagram showing that in green our chelating agent has fully encapsulated these metal ions, the chromium. So in conclusion, I'll just point out the chelates can stabilize very, very reducing metal centers. These metals are amongst the most charged, reducing things to ever exist in water and not make hydrogen. And that's been sort of our in on getting to these very, very high performance metrics. The materials we're using iron, chromium and these chelating agents are scalable and have domestic supply chains. There are plants in the United States cranking out chelating agents for the paper pulping industry, for agricultural applications, for consumer products and food. These chemistries solve new battery challenges. I wanna point out here that this is not engineering a new chemistry, or sorry, this is not engineering a chemistry that's 30 years old. This is a brand new chemistry that we're using in batteries for the first time. Our 1.62 volt cell potential is what really drives this high power and high efficiency. Based on some technical economic modeling by a third party conservatively, we're getting below $50 a kilowatt hour. Potentially as we scale, we can get even better cost metrics, especially since these were based on not taking advantage of things like lower cost membranes or other advantages of our system. The systems able to achieve the levelized cost of storage of five cents per kilowatt hour of the DOE Energy Storage Grand Challenge, I would point out, we have not yet seen, even looking very carefully, we have not seen material degradation. So we still are looking for it, but it seems like it's at least going to last for 10, possibly 20, possibly longer. Can't say that. And then as Will mentioned at the beginning of this talk, I co-founded a company, Otoro Energy, along with Dr. Tyler Evans, who co-founded Salion and we're leading a $4.14 million contract from the Department of Energy's Advanced Manufacturing Office. So we're taking this chemistry out of the lab and we're gonna scale it. And so I'd like to thank here, my group. I've had a number of questions and I'd like to thank here, my group. I've had a number of talented postdocs. I'd specifically like to point out Dr. Brian Rob, who recently graduated and is now employing number one at Otoro Energy, along with collaborators at Raytheon, Otoro and Sandia National Labs, along with a number of talented students who've gone on to do some great things. So thank you again for listening to my talk and I'm excited to hear feedback. All right, Michael, thank you so much for sharing the latest and greatest results. They look extremely interesting and promising. We have quite a few questions, but we're running a little bit short on time. So let me just pick two questions and they're both specific. So you showed a lot of cycling results, Michael. Can you share a little bit on the calendar aging at the full state of charge? Yeah, we have done studies where we've charged up the electrolyte not to 100%, but to 90 to 95% and then stored it in a glove box and just seen capacity retention. And what we see is that there's a self-discharge rate of I think it was 3% over the course of one month and that 3% was a self-discharge as a result of hydrogen evolution. So we're not seeing any material decomposition. What we're seeing is a slow reversion from the charged chromium-2 state to chromium-3 along with hydrogen evolution. What we've kind of observed over time is that the rate of this decay and also the current efficiency that we observe in our cell is highly dependent on the purity of our material and specifically the presence of certain impurities. There are obviously metals like copper, nickel, obviously platinum that on contact with our electrolyte will rapidly catalyze hydrogen evolution at these voltages. And so one of the challenges that we've been working on is efficient ways of removing these impurities to prevent hydrogen evolution. But what we think moving forward is the fact that we were able to isolate this and crystallize it and put it in a bottle. Suggest there's no intrinsic instability of our molecule itself. And again, we haven't seen any material degradation. The chelating agent does not fall off. It's got a high binding constant and the chelating agent doesn't decompose. What we're seeing again is just this catalytic hydrogen evolution from various impurities that are present. Michael, it's like you can read the minds of our audience. The second question I was gonna ask on behalf of our audience is impurity. The reason I think this is important is a lot of the CO2 footprint in lithium-ion battery comes from the need for very high purity materials and the separation and refining process. Typically even for lithium sources, you will need three nives and above for lithium to achieve the needed performance. And you really have to pay a price for this, especially in terms of cost in CO2 footprint. Michael, so can you maybe give a little bit more insight on what is the purity level that is required, as you mentioned, to completely suppress HER? And whether or not this might become a cost-limiting factor. I'm a huge fan of chromium actually. It's one of the more abundant redox active elements out there. And this really I think underutilized as a redox couple. So I'd love to hear a little bit more about what is needed in terms of the purity. Well, one thing I would argue from the beginning is that the premise is not necessarily correct. We don't need to completely eliminate hydrogen. Remember that we're not operating in a crimped lithium-ion battery that you can't open up. That this is something where hydrogen that comes off the system can be fed back in to maintain a charge balance of the overall system. You can think of a flow battery, not as a battery, but more as a chemical plant that's taking electricity and performing two chemical reactions. And yes, it can get those electrons back out again and it can act as a battery. But because you're storing these liquids and tanks, it's much more like a large scale chemical reactor. And so because of that, you can think about, well, if you had hydrogen in your head space, well, is that really a problem? Could you feed that back in to the system and rebalance it? I think there's a number of battery, flow battery technologies, including vanadium and including all iron systems like what ESS is doing. They've talked a lot about their hydrogen rebalance systems. And so I think that hydrogen is not necessarily a massive problem. It does reduce your efficiency, but there's a trade-off between purity and efficiency here. And we're not irreversibly losing material because of these inefficiencies. Now, to get to how to purify, I think there are a number of ways that we are looking at doing that. I would point out also that the grade of things like vanadium in vanadium flow batteries has to be purified beyond what is typically coming out of a mine. So the purification I don't see as a really significant challenge, again, because we have a trade-off between efficiency and purity that we can play with in terms of optimizing our costs. Great, Michael. I think we can discuss more in the panel discussion. But thank you very much for sharing these great results and good luck with your talk. Sounds extremely exciting. So maybe let me ask Yi and Fig to come back and we have about 20 minutes and we can have a spirited discussion. Yi, would you like to kick us off? Sure. Maybe I'll just continue. I want to ask Michael a question using this opportunity before all discuss with Fig and Michael together. Mike, a very interesting molecule is the EDTA and PDTA. So one question I was wondering because the potential is so low once it compresses with chromium, it's exceeding the hydrogen evolution potential by so much, but you can still stabilize that. And then you could still do the transfer oxidation and reduction with the electro with a decent weight capability. That means this electron can go into chromium even with the PDTA. With high rate. So this seems to be in conflict with the stability like as water cannot access the proton, the hydrogen cannot access to these chromium. A very interesting question. I want to pick your insight about this in thinking about this problem. Yeah, that's a great point. So what I think we're taking advantage of here is the fact that the Keelan agent can protect the water from chemically reacting or getting close enough to chemically react, but it doesn't, it's not so big that an electron can't tunnel through it. So that these materials, these metal complexes can get close enough to the electrode. And in fact, the coordination uses a carboxylic acid, something that's conjugated so that perhaps the outer oxygen could have an internal electrical contact with the electrode as well as the metal center itself. So I don't know that that's what's going on, but what I would say is that the Keelan agents are probably more chemically inert, but they may not be perfectly electrochemically inert. That we can get electrons in and out like you mentioned without getting the chemical reactivity. Yeah, so very interesting. Okay, I'll ask a question to both of you, Michael and Fik. And Fik gave this analysis over really multi-parameters. You got to look at a redox flow chemistry, the whole system. Now using these criteria of what Fik presented, Mike, you know, you show this exciting data. And it's a Fik, if you look at the Mike's chemistry, very interesting by using your multi-parameters analysis right here, has to say that it's more like a Fik, I asked Mike question, now I asked my question using Fik's perspective. Mike, do you want to answer that? Or Fik, do you want to also analyzing Mike's results a little bit using the criterias you put together? And what's your thought there? So I have not analyzed Mike's data, but I know smart people at Raytheon and third party folks have done it too. I would say that our techno-economic model has been published for a number of years so people can pull it down and apply it to their own chemistries. And I would argue that they're probably more advanced techno-economic models that exist that take into account a lot finer detail than we went into in that model. All that to say I'm excited about Mike's chemistry. I think there's lots of really interesting opportunities. And I think one of the points that he made that I maybe want to reiterate is that we want to have new ideas. We need to have new chemistries. We need to sort of try to think about flow batteries in maybe a slightly different way than we might think about lithium ion batteries in terms of the way that they operate and the terms of the way that you think they're economics. I think thinking of them as a chemical plant rather than sort of a portable energy storage device is a change in sort of vantage point of thinking that we need to go through. And I think being able to communicate that and to show exciting new chemistries that really push the boundary of what we thought was possible will hopefully generate more interest in this field, more investment in this field and ultimately drive down the cost for everyone. So I'm excited. That's great. Will, do you want to take back or do you want me to ask one more questions which I do have one more question? Please. So this is good. So Mike, I mean, looking at this PDTA, so the annual side, the negative electro can go down. The reason seems to be PDTA complex with this molecule or EDTA to favor the, I would say high oxidation state more. So to make it harder to get rid of that, the potential is lower. If I would take this the same thinking to say, well, what about a catholic? Can you bump out a catholic redox potential? And by doing complex, can you share with us some of the thought on the catholic path? So that's a great point. That's something we've looked into. What I didn't talk about today, we've done this with iron. However, it pushes the iron redox couple down. Kylation typically pushes everything down more negative. And we can take advantage of that by taking redox couples that normally would not be accessible or normally would generate hydrogen or cause some sort of decomposition and then bring those down into the water operating window to still get out high voltage. However, we see a lot less decomposition. So one thing we're working on right now in the last, collecting the last data is looking at a chelated cerium complex. And in that case, we've demonstrated cycling of a 1.9 volt battery where the chelated cerium complex is just outside the 1.23 volt water window on along with the chromium complexes. And now we have two coordination complexes that are both outside the water operating window. That said, I think when you get to start, when you start to move towards two oxidizing a molecule, you run into problems not only with oxizing or chelating agent, but you start to corrode things like the carbon electrode itself and you start to lose compatibility with hydrocarbon based membranes and other components. And so I think there's a lot more of a limitation on the positive side in terms of oxidative corrosion. On the negative side, I think we can continue to try to push that even further. And that's another area that I'm really excited to start looking into as well. Yeah, me too. And also based on this, you're starting to come back to fixed analysis. So this voltage right there is a denominator, very powerful to reduce the cost. So any voltage gain you have right there can means a lot, yeah. Well, back to you Will. Thank you, Yi. Well, maybe building on this theme of connecting the two work together, you know, I also am very curious about this notion of viewing flow batteries as a chemical plant. And I think, Michael, you already alluded to this that the hydrogen in the Hespace actually can be recovered and also combined with additional energy cycles similar to topping and bottoming cycle in a mechanical thermal system. So my question is, you know, ultimately you don't wanna have too much hydrogen otherwise you can just make hydrogen directly as a storage medium. What is that happy balancing point in terms of parasitic current and the self-discharge rate, right? So for lithium ion, the self-discharge rate is extremely low because you can't afford to recover any of the parasitics as being one limit. And then you have system in which the parasitics too high even if you can recover it, it doesn't make sense overall. So what is this happy balance? Maybe I can aspect the comment on this first and then the prospect of incorporating some sort of a recovery system to improve overall energy efficiency. I'm happy to go first. I think though Mike might wanna say a few words and I can follow up behind him because he's on the front lines here but Mike, if you want, I can go first. Sure. All right, so, well, it's a really good question. And it's one of kind of capital expenditure and sort of operating expenditure that is to say as Professor Marshak pointed out, right? If you need to worry about the purity of your electrolytes for every additional nine that you're going to it's gonna cost you a certain amount of money, right? And then you're sort of asking yourself what am I buying with that extra nine in terms of let's say reduction inside reactions? And really what it becomes is now a question asked to, well, how much is low enough that the cost of say an additional unit to recover the hydrogen and re-inject it becomes overly expensive or not that expensive at all. So I suspect that it will be sort of a, there's a trade-off but that trade-off will be somewhat chemistry specific. Just as an example, I think it was mentioned earlier for the all iron system that has a fairly high hydrogen generation rate on the iron plating electrode going from iron to iron metal but they can essentially re-inject that into the system using an iron hydrogen fuel cell which is built in there. And they've probably done the economics and the economics tells them that you could have a certain hydrogen generation rate and they have a certain hydrogen recovery rate and that makes the whole system work. So I think that what it's ultimately going to be is a balance. And the one thing we aren't talking about here and that's on purpose is how are you actually using the battery? Cause that could also be a differentiating, defining factor where having to sort of re-inject that hydrogen may impact the application space you can use the complexity of being able to use the system in real time. And so that would also probably need to be factored in. I think that covered most of the angles but you may feel differently, Professor Marshak. No, I think that was a great summary. The one other piece that I would add is that the other important piece other than efficiency you have to worry about with hydrogen is safety. Is that you're dealing with the flammable gas and ideally you can recombine it in real time. And that's how a lot of these recombination type systems work but the recombination in real time essentially eliminates any sort of flammability hazards that would be associated with it. However, if you don't do that or if you allow it to build up especially building up pressure can create real safety hazards and that can limit the siting of where you put these types of batteries. One thing I'm really optimistic about with our technology is the fact that we're relatively non-toxic, pH neutral and non-flammable is that we may be able to put these types of batteries in places that a typical lithium system might not be able to go because of the flammability hazards and other pieces. Can I pick up on that just for a second? So I think another thing kind of perhaps coupling to that is it was mentioned earlier that energy density is not a key driver for grid scale energy storage, that's correct but energy density can unlock some application spaces. And so the nice thing potentially about this chemistry is because you're high voltage you could have a smaller installation footprint coupled with the safety aspects it means that the potential for urban installations rather than remote installations becomes a little bit more interesting and your actual installation footprint gets smaller which is also an attractive feature for being able to site systems. So they all start to connect to one another a little bit. This is great discussion, Fika and Michael. I just wanna maybe, I wanna resonate with the point you're raising here is energy storage is all about trade-offs and you can't have everything but you have the things that make sense. And one future I'm imagining is that maybe you accept some parasitic losses but you get to use very low purity materials and with chromium coming from a mine, minimum separation and then you dump it into a flow battery. And this is really what's needed to create the hundreds of terawatt hour of storage that we need. But I think to do so we have to give up something and I think a system level analysis will reveal where the sweet spot might be. So very excited to see the work that you're doing at the systems level effect to understand the co-optimization of so many things. I think looking at battery is a great example. It doesn't have everything but it has the important things that matters to the use cases. I wanna come back just really quickly before handing off to E on this topic of energy density. That's also another major trade-offs, right? Flow batteries have low energy density, larger deployment footprints. So if I can also ask maybe the both of you to muse on what is that optimal balance? You're getting this decoupling of energy and power that's what you get. You're getting perhaps a more scalable chemistry than other technology like lithium ion but you paid the price of energy density and having a larger deployment. So where does that optimum apply? And how close are we to it for redox flow batteries and what are the additional innovation needed to get to the sweet spot? I'm happy to go first. So the reason we're quiet is because it's a difficult question to answer. And the reason it's a difficult question to answer is that it really depends upon a bunch of factors that are unique to each installation, the value of land, the availability of it, things of that nature. I would say that sort of I finished up the last answer with urban installations which might be desirable for essentially supporting large load centers, right? That's gonna be a little bit tricky because you only have the cost, the price of land is significantly higher because everyone wants to live there and you have to worry significantly more about sort of safety and what happens if you have an event. If you go out to a field in the middle of nowhere, then of course you don't have to worry as much about installation costs because the price of land is a little bit lower potentially. But I do think that there has been innovations in the field. One of the things I think, I don't remember if I showed this, I don't think I did, but some of the work around containerized solutions where flow batteries are sort of built into shipping container type apparatus can be very helpful, not only for sort of plusing up existing installations, but also for being able to move it around and take advantage of the way that we've moved things around the country and around the world for many, many years. And so I think there will be some benefit in sort of taking advantage of building the system in a way that can most easily fit into the application space. I wanted to make one comment that's a little bit off topic, but I think a line to what you're saying too will. And that is, if you think about a flow battery as a chemical plant, you can think of it as you can start off with one chemistry, right? Could be the installation is with vanadium as a front-up chemistry, cause it's available. But if your operating life is long enough, at some point in time, you could drain the vanadium and put in another chemistry, right? It's of course, it's not that simple, right? There's a ton of technical details, but on the high level, you can imagine that if you're building that infrastructure, there's lots of different sort of processes that you could add on there. You could begin to augment the system adding new chemistries, right? So that's what makes me very excited about this, is that you might put in an installation and you might have it as a certain footprint, but let's suppose we start up with vanadium and then we put in some of Professor Marchek's chemistry, you know, let's say five, 10 years down the road and all of a sudden we've augmented the amount of power that we can get out of that system, right? And so I think that's a very interesting way to begin to think about these types of systems in the long run. Thanks, Mike. Michael. That's a great, that's actually an excellent point, Vic. And as you know, one thing that's really accelerated the development, both in my lab and as well as with our collaborators at Raytheon was this chemistry really being a drop in replacement for vanadium as a starting point. And so the fact that it works pretty well that way and then we can further optimize and tune beyond that is great. And so I guess one way I think about it is sort of the, there's, you know, when Tesla got started, for example, they didn't build a car from the ground up. They worked with Lotus Elise, Chassis and some laptop batteries. And as they've evolved as a company, they've kind of re-engineered each component to eke out more efficiency, more performance, et cetera. And I think that that's something that really provides a compelling pathway forward with our technology is that we can already leverage the 30 plus years of R&D that's gone into vanadium flow battery development and then be able to kind of go back and tune each component further to get out better performance and better efficiency further. In terms of the siting, I think energy density, like Vic said, it's very site dependent. But one thing I would point out is that a lot of the early markets that I think are gonna pop up for this long duration storage are going to be places that are operating critical infrastructure. And we're thinking, whether it's a data center, a hospital, places that can't afford to lose power. And you can't cite backup power generation far away because further away your backup power is the more risky run with the power transmission system. And so if you look at a hospital, you're gonna have onsite diesel right there. If you look at many of these other critical infrastructure places that they have onsite power generation because that's where they need it to be. And I think that energy density and looking not just at the energy density of a lithium ion cell chemistry, but looking at the overall system footprint per megawatt hour or per possibly gigawatt hour is the much more accurate comparison to be making here. And if you can build these aqueous systems in a more vertical footprint, then that might be a compelling way to address the lower energy density comparatively of these systems. Micah, your comment hit too close to home. Stanford recently lost electricity for four days due to a wildfire. So could have used your battery then, B? Yeah, I want to ask Fik and Mike. This is very nice discussion. Let's imagine the blue sky. Extremely long duration energy storage. Well, what's your thought on how do we get to $10 per kilowatt hour or less? Let's just imagine, you know, and what does it take to get there with a redox flow? And get to that level, you need to pay attention to every single detail, every single component, that's for sure. And can you share with us, you know, do you see a path like how do we get there? Fik, do you want to start or should I? I'm happy to take the first crack. Okay. It's tricky, but I think what it basically boils down to is your floor is chemical cost, right? And then everything is additive on top of that. So you want to start off by saying, what chemicals can you pull out of the ground or wherever they're coming from, that will lead you to a chemical cost of, you know, when combined together $10 or lower per kilowatt hour or something like that, right? Then of course, there's the add ups on top of that. There is, well, what's the purification going to look like? What is the efficiency of the battery operation? Things of that nature, but those are all plus ups. So if you're not there with the base chemistry costs, it doesn't go down, it goes up, right? I think what it will end up by being, and I think this was mentioned earlier, is that you're going to have to have systems that are relatively poor in power performance. But what you're giving up in power performance, you're saving in total system cost as in dirty materials, dirty as in like, we're not purifying to the end nines, right? A cell geometry that is just very, very, very simple, right? You know, it can be manufactured without significant precision machining. And then probably an operation where you might not be that energetically efficient, but you're relying on the fact that your costs of energy are just so low that maybe you can get away with it. So one approach that's kind of going that way is kind of what they're thinking about in form energy, where they have iron metal, which is sort of dirt cheap or an iron oxide and manufactured domestically, and they're relying on air for the counter reaction. And then essentially the game that they're playing is, how little can we add on top of those base costs in order to get us to this really low system cost? But what you do when you play that game is, it's not necessarily the most efficient system or the most powerful system, but it's for these longer duration applications. So that's a hypothesis as to how you might be able to get there. There are other ways to do it, but this is kind of thinking about purely the flow battery space. Yeah, good, Michael. I would start by saying, I mean, I agree with all that. And I think as a chemist, I'm a little bit more cynical about the possibility of getting to $10 per kilowatt hour. And part of the reason for that, I don't think that the trade-off with efficiency is really a good ethical decision. I think if you're running a battery that's less than 50% round trip efficient, then even if you assume the grid is 100% renewable, you're still having to provide an overcapacity of wind or solar that has a carbon footprint associated with it. That it's not an ethically favorable proposition to develop a system that burns or effectively loses half of the energy that you put in it every time you cycle it. And the other thing I would point to is that as you start to get to longer and longer durations, I have great optimism that there are gonna be emerging demand response systems that will be able to accommodate excess power on the grid. For example, things like electrification of steel production, the electrical CO2 capture and conversion or storage, things like that where if there's an overabundance of electricity in the summertime, for example, with the longer days, we can do those processes for part of the year and then curtail them in the wintertime and just use the power we're generating for our other needs. So I think that there are gonna be, but as you start to look for longer and longer duration applications, you run into this challenge of you're coming up against other demand response systems. People can just turn out the lights, shut down the factories for a week. There's a lot of different other approaches that start to become much more favorable than trying to make a chemistry that's very, very low cost and very poor efficiency work. But I mean, I'm excited to see, I think we need to take an all of the above approach on all of these, and that's for sure. 100%. Very nice, yeah. Back to you, Leo. Absolutely agreed. And it pains me to have to end the session it's such a great discussion, but I will ask one last question. So with all decarbonization, decarbonization pathway, there is a need for scale and speed, right? So each touch upon scale with costs and deployment. Maybe I can get from you, you know, one minute each speed. You know, how do we get this to actually deploy in a reasonable amount of time? What can we do? What is needed? Where are the opportunities to speed up? So I've talked a little bit too much. So maybe I'll let Mike go first. Sure, I have to leave in about five minutes to teach, but what I would say is that there needs to be funding, and I would say federal support at all levels. I developed this technology using startup funds from CU Boulder because I wasn't able to apply and get energy storage funds. And this is a much larger problem, I think, but there is no office of energy storage at the DOE, like there is for solar energy or for wind or for energy efficiency. There's really not very many open calls. I was looking recently, even with the energy storage grand challenge, there are no topics in the SBIR program for energy storage right there right now. Like it's crazy that there needs to be a much more concerted and organized push in energy storage at all levels, whether it's academic or the demonstration and deployment level. And those need to not happen every three years, they need to happen every six months. I like the sound of that. So I would echo what Professor Marshak said, and I would maybe put another sort of spin on it, is that the markets for long duration energy storage have yet to fully develop. A lot of times the reason you're not seeing so many full batteries on the grid is because there isn't necessarily a need for them. And so the economic justification doesn't exist yet, but we know that it's coming. And so you have this problem where you're trying to build heavy duty electrochemical machinery for the future that requires years of development, testing to go from a prototype to a product and to be scaled. But we need to find ways to sort of smooth that pathway, as Mike said, from the research lab all the way up to demonstrating and testing and de-risking these types of technologies. And I think this is a really great role, potentially for government funding, public-private partnerships. And so I think just a little bit more connectivity from the lab to, we'll say, the real world, the industrial world, I think will be really helpful, especially in terms of understanding what are the main concerns for power producers, what are the things that they prioritize and whether some of the things that we think are an issue as academics may not be such an issue for them. And some of the things that we're glossing over may end up by being real showstoppers for them. And so the more we talk, I think the better we get to powerful solutions and the more we can take advantage of the intellectual horsepower around the country and around the world. Right, Mike and Michael and Vic, really appreciate your comments, the need for skill and speed. That will be the model for the next 25 to 50 years. With that, I'd like to thank you both once again for joining us on this really great discussion and presentations on Redox Flow Batteries. Kaylee, if I can have the closing slides please. Thank you. So we have two more talks coming up this summer. The next one is also on grid-level energy storage, looking at aqueous energy storage with sunroof energy from the city college in New York and WG from Oregon State. And then two weeks following that, we will go to material informatics as a way to accelerate energy storage R&D. And with that, I'd like to thank everyone again. Please stay connected with us on social media and look forward to seeing you next time. Thank you so much.