Energy storage is heating up

Kara Miller: I’m Kara Miller.

Robert Stoner: And I’m Rob Stoner.

KM: And from the MIT Energy Initiative, this is What if it works?, a podcast looking at the energy solutions for climate change.

Today, a scientist who knows that coming up with tech that’s going to rock the world is not easy.

Asegun Henry: If you can’t beat the existing state-of-the-art, you don’t get to play. Full stop. And so, you’ve got to offer something above and beyond what exists today, otherwise you should go home.

KM: And remember, says Asegun Henry, rethinking energy tech and rethinking tech we might be a little more familiar with, those are two very different things.

AH: You could look at, for example, iPhones as a consumer technology. And you look at how quickly iPhones took over the market. It’s very enticing to want to think that energy technologies could proliferate that quickly. And, you know, it’s the right solution and so everybody’s got it a year later. It doesn’t work that way.

KM: Henry argues that one place where we definitely need something beyond what exists today is in energy storage. And even though there are lots of competitors in that space, it may not always be that way:

AH: There likely will be a single or very small handful of approaches that end up really dominating because they meet the needs the best and many other things will fizzle out. And so I’ve put so much time and energy into Fourth Powerbecause I think it’s I think it’s going to be one of these key winners that ends up out there in the end.

KM: Asegun Henry believes that when people imagine what the energy transition looks like, they frequently make a central mistake. They forget how much this is a physical problem—a problem that, largely, is about big, heavy, stuff, scattered all over the world.

AH: When we think about the future energy system and what it will take to get from where we are today to what the future energy system will look like, there’s a lot of new hardware. This is not a software problem, although there may be software solutions that help along the way that make things more efficient, that do things better and more safely, etc. At the end of the day, there is a need to physically change the infrastructure that we use to convert energy from one form to another and deliver it where it’s going to be used.

KM: Henry notes, all you have to do is glance back at the past to see how this sort of transition has played out before.

AH: When you look at the various histories of different technologies and look at the time scale it has taken for humanity to switch fuel sources, like from wood to coal and from coal and natural gas, it’s typically a 50- to 100-year time scale. And developing a new set of hardware is also an order of 10-year time scale to just develop the technology itself.

KM: But if you don’t feel like we have got a half century or more to figure this out, what do you do? Henry says, relying purely on capitalism may not do the trick.

AH: If you’re waiting on the natural driving forces in a market to take up a new technology and start using it, that time scale may be far too long, much longer than we have. And so, at the end of the day, what I think my perspective is on the field is, we only get pretty much one shot.

KM: So how do you take that one shot? Henry says, you’ve got to go big.

AH: We basically have to parallelize everything and try every idea that we can think of that has merit and try to get to whatever the cheapest, most effective technologies are as fast as possible so that we could minimize the total cost of deployment. The deployment cost is really where all the cost is. And so, you know, it’s like, why not spend an extra 1% to save 90%?

KM: And there are examples of technologies where investment resulted in plummeting costs, like wind for example, or photovoltaics to capture solar energy.

AH: So those are great examples where you have a technology that call it 30, 40 years ago was kind of laughably expensive compared to the existing infrastructure, but over time has now overtaken in a sense and become one of the cheapest sources of electricity on earth. It’s one great example where we got 10X from kind of economies of scale and sequential iterative development. For, I think, storage is another place where there are ideas. Of course my company’s pursuing an idea—Fourth Power—that is 10X cheaper than the current state-of-the-art.

And I think this is an area where you see potential options, where, as Bill Gates liked to call it, the green premium, meaning how much more do you spend to get the clean technology compared to the existing technology? This is one of the really rare cases where you have an opportunity to possibly get cheaper than what we use today, like natural gas. Where if you pair these really cheap energy storage technologies, such as the one we’re pursuing at Fourth Power, if you pair that with renewables, you get to essentially a high-capacity-factor, fully dispatchable, renewable energy power plant. That is the same or lower cost of natural gas.

RS: And so what’s the technology?

AH: Oh, how does it work? Yeah, so the key point here with Fourth Power’s technology is we made a pretty radical departure from normal energy storage. So, when you say the term “battery” typically people mean an electrochemical battery, which is one that has an electro-chemical reaction going on inside. What we decided to do differently is instead of storing energy electrochemically, we store it thermally. And the premise of this is that storing energy thermally is about 10 to 100 times cheaper than storing it electrochemically—just from a fundamental basis.

So, we started by storing the energy in a different way. There’s a big trade-off you make though when you store energy thermally, is that you then need to convert it back to electricity. And when you do that, you lose, call it, roughly speaking, about half of it. And so you end up wasting some energy.

And for a long time, this concept of going from electricity to heat, back to electricity, would have sounded absolutely stupid. And the reason it would have sounded stupid is because we go through great pains already in the current energy infrastructure to convert heat to electricity the first time. And when we do that, we lose more than half of it. And that’s a big waste.

And so we put great effort into trying to retain and convert as much of that energy as possible to the final form we want, which is electricity. So the idea that you’re gonna now go back to heat only to pay the penalty a second time, would sound like, it’s so bad, we used to call it a thermodynamic crime. Actually, that it’s a crime against thermodynamics to go back to heat only to do it to pay a penalty a 2nd time. But there’s one main reason to do it, and the only reason to do it, is because of cost. Meaning, storing energy thermally is so cheap that it actually makes sense to waste some of the energy along the way and pay a factor of two hit on efficiency to get access to an order of magnitude lower cost.

KM: So, explain this to a non-scientist for a minute. You reference the fact that generally, initially when we get electricity, there’s a creation of heat. And I assume that could be like a coal-powered plant would be that, right?

AH: Yep.

KM: You’re getting heating things to very high temperatures and out of that, you’re powering the plant and powering your city or your town, right?

AH: Yep.

KM: Okay, so in what you’re doing… What is it that you’re heating up? I assume things are going to very high temperatures. What is getting heated up? What does that look like? And should people be concerned about like that level of heat and…?

AH: Yeah, so to the first question of what are we heating up, what we’re heating up is graphite blocks. So, graphite is essentially, you can think of it like pencil lead.

KM: That’s how I think of it.

AH: And it’s a very inexpensive, yet very refractory material. When we say refractory, it means that its maximum temperature that it can remain solid is a very, very high number, very, very high temperature.

KM: Okay, so it kind of still looks like graphite. It’s not melting all over the table sort of thing. No.

AH: Okay. Yeah. At atmospheric pressure, carbon doesn’t form a liquid, so it just sublimates when it gets hot enough. And so graphite can be used beyond 3,000 degrees Celsius. And so that is the main thing that we’re storing the energy in. And what’s unique about the system we’re developing and commercializing at Fourth Power is that there’s another material we use that allows that system to have some really unique features, which is liquid tin. So we use liquid metal. Which is essentially the same metal you see in solder. I don’t know if you’re familiar with soldering.

KM: Sure, sure, sure.

AH: If you solder two wires together, that liquid, it melts at a very low temperature, the same temperature you can get a soldering iron to. But what’s amazing about that particular element on the periodic table is its boiling point is 2,600 degrees Celsius. So it has a huge range where it remains liquid and doesn’t boil. And so we exploit that.

And so, we use that liquid to actually move heat around in our system. And what it allows us to do ultimately is two major things. Number one, it allows to decouple energy and power. So a typical, let’s say lithium-ion battery, it’s kind of made like a fruit rollup of materials that want to interact with a separator in between. In our case, when we’ve separated power and energy, what I mean by that is our system has separate components that do the conversion of energy from one form to another. And it has one component that is storing the energy, which is the graphite blocks. So if you now—and these things are physically located in different places—so if now wanted to increase the amount of energy you’re storing, all you have to do is keep adding more graphite blocks to this big, very large structure we have of graphite blocks. And so you can do that independently of changing any of the power hardware. And this is really attractive for utilities.

Utilities are our customers at Forth Power. And it’s really attractive for them because they have to size the energy storage that they wanna use on their grid. Maybe they want a battery that can output for 10 hours. It’s projected in the future, their needs are gonna evolve and they’re gonna need 20 hours and 30 hours and 40 hours. And the duration of the need of discharge is gonna extend as more renewables and other things come onto the grid.

And so we have the ability to actually grow with the grid and evolve with the grade where we can install an initial Fourth Power thermal battery. Let’s say it’s a 10-hour battery. And later the utility could decide, you know what, we really wish that was a 20-hour battery. And you can augment it and retrofit it to just add more storage. And the addition of more storage is only like 20% of the cost, the original cost. So it’s a very small increment to now keep adding on and tacking on more storage to change the duration. So that’s one of the key features.

The second key feature with the liquid metal is power density. So it allows us to shrink the total amount of hardware you need in order to make a system of a given power rating. And ultimately what this does is this cuts the cost low. And so this is how we get our costs really far down is by leveraging the liquid medal as a medium that can transfer a lot of heat with very little equipment.

RS: So you’re saying basically you just put more metal in the pot.

AH: I want to be careful because you said metal. So we’re using metal to transfer heat, but you put more graphite blocks in the pile in order to get more energy storage.

KM:  And then how much of a game changer, there’s sort of two twin questions. One is, how much a gamechanger do you feel like this is? And then, how hard slash, I mean, you’re obviously a professor, you’re kind of an entrepreneur here too, how hard or easy is it to launch this in a big way to get it to be widespread?

AH: Our techno-economic models show that this is, quote-unquote, the game changer in terms of being able to proliferate storage onto the grid in a cost-effective way that will allow us to transform the grid and decarbonize very, very deeply, if not fully. It’s a platform technology. You can do multiple things with it. So taking in electricity and outputting electricity is the number one target market we have for starters. But one of the amazing things you can also do with it is you can take in electricity and output heat. And so heat is another application where there’s big energy needs. And because our temperature is so high, we can basically service any application on Earth that needs heat.

In addition to that, there’s another feature where you can actually bring in fuel. And because you can react fuel to give off heat, you can actually charge it up with fuel rather than electricity. And so this third way that you can utilize it allows you to really use it as an ultimate backup for the grid and keep the grid going kind of under all circumstances. So from that perspective, we think it’s an absolute game changer right now. We are in the process of, we’ve built some very large scale accelerated-life test rigs at a facility we’re working out of.

RS: Meaning rigs that they’re used to test how long these things will live.

AH: Yeah, I call them rigs because they are not the full system, fully integrated. They are subsections or subsystems of the full system that we are testing for longevity and operability. And then the second thing that we’re doing is next year we’re building a fully integrated one-megawatt-hour system at the same location. And so once we do that we expect that that’ll be a very large value inflection for the company to show all this full-scale hardware working together. That’s, I would say, the primary challenge in front of us right now. And it’s largely one of, I would say, timeline because we built the entire technology in my lab before we spun it out, before I spun it. And so I’m confident that it works. It’s not really going to be a surprise that it works. What’s really the new thing to demonstrate is that it works when you now scale it up, meaning the components that we’re using are not small anymore. Now they’ve got to be lifted and hoisted into place with cranes. And these are the same kinds of dimensions you’re gonna see in the full-scale product. And so you de-risk a lot of the risk associated with scaling up all in this one demonstration we’re doing next year.

RS: So the temperatures are operating, what sort of efficiency are you getting from the device? And how does that compare with the efficiencies of batteries that people are developing that they also claim can be very low cost and very large scale? Things like molten metals and so on.

AH: Yeah, so most of the technologies that I’m aware of, that claim anything even in the realm or approaching the costs that we claim, are all making the same trade, which is they get to lower cost by trading efficiency. So typical lithium-ion batteries are greater than 90% efficient and all the battery technologies that get to like well below $100/a kilowatt hour are all trading efficiency and are closer to like 50% efficiency, or 60% in that range. In our case, for our Fourth Power thermal battery, the target is to get to 50%. And all of this loss is really dictated.

I didn’t actually finish explaining how it works, but the entire system is extremely hot. So the nominal operating temperatures are between 1,900 degrees Celsius and 2,400 degrees Celsius. To put that in perspective, 2,400 degrees Celsius is about the same temperature as an incandescent light bulb. It’s about half the temperature of the sun. So there was an earlier question you asked about safety. So that’s usually the example that I give for people, like everybody’s actually been a couple inches away from something at 2,400 C, which is a light bulb filament. So high temperature inherently on its own is actually not dangerous. Typically, what is dangerous is high pressure, which is what can lead to explosions and mechanical failure. You know, what makes a light bulb safe is the fact that it’s got the bulb around it, right? So the point is that there is a barrier between your hand being able to try to touch something at 2,400C. And the same is true for our system. There’s an entire warehouse built around the system that prevents anyone from being able to walk up and touch or even see any of this glowing white hardware. And this is where the moniker came in the news calling it a “sun in a box.”

KM: You’ve talked about the notion that climate change itself is a big enough problem, and also, the sort of transition that has to happen is a big enough problem that it’s really government-scale money. Now, in the U.S. currently, we don’t really see the government investing huge amounts of money. So how is the private sector doing? I mean, that’s, I assume, who’s coming in and saying like, okay, let’s take a look at these batteries. Let’s implement them. How’s that going?

AH: It’s slow. And I think this is, I think, a principle, I call it a fundamental problem that we have, which is I think there is an assumption on the part of, you know, we look at different technologies and their proliferation, particularly things that don’t require a lot of hardware. You could look at, for example, an iPhone as a consumer technology and you look at how quickly iPhones took over the market and it’s very enticing to want to think that energy technologies could proliferate that quickly and it’s the right solution so everybody’s got it a year later. It doesn’t work that way. The physical amount of hardware that has to be deployed is so large the time scale is just longer.

And so the main concern I have about the rate at which we are moving is I think that there’s an assumption built into the idea that you’re just going to build some cheaper solution and it’s going to get proliferated fast enough. And that assumption is that the rate of which business or the rate at which the market will take up a new technology that even at its maximum speed is actually fast enough to meet the goals that we have for the it. And I think that that assumption may not be correct, that in order for things to move fast enough, there has to be like an external driving force, something else that causes things to happen much faster than they normally would. And I generally label forces of that nature, governmental in nature, meaning it’s gotta be a collective decision on behalf of humanity to actually take some steps that will actually either de-incentivize the emission of CO₂ or promote its replacement so that you can accelerate its decommissioning.

And this is actually an important I wanna point out… I think I’ve probably said this publicly before, but I love to highlight this point because I think there’s a big misnomer. Building renewable energy on its own actually doesn’t solve climate change. Stopping the use of fossil fuels is what stops climate change. And those are two actually very different things. You know, I’ll use the analogy of, let’s say, someone that is physically unhealthy for eating a lot of unhealthy foods, right? And it’s like you add some salad into their diet and just adding salad doesn’t actually keep them from being unhealthy if they still eat the unhealthy food, right. So at the end of the day, we have to somehow either de-incentivize the emission of CO2 or incentivize things that can decommission those assets quickly. And that’s actually what stops climate change, is stopping the emissions of CO₂.

So to your point, I think that it’s going slow. I think that if you talk to folks that are part of the Intergovernmental Panel on Climate Change, they’re watching the problem closely and we have not yet hit the turnover. We have not yet decreased the amount of CO₂ we’re emitting. We’re still increasing every year. Business as usual is still continuing and it’s still getting worse. Now, here’s the key point just to tie this back to what I just said. The amount of renewables on the grid is going up. There’s actually been a significant amount of renewable deployment and yet the CO₂ emissions are still going up.

KM: I think China is a great example of that, where they build coal-fired power plants and renewables are taking off. Like two things can be true at the same time. It’s not like you have to choose, which obviously is not good for climate change, but solar power can be spreading like crazy, and you can be building coal-fired plants, if you’re industrializing and so on.

AH: Absolutely. And this to me just further underscores the main point I want to make about storage being arguably the single most important technological problem we have to solve. Because storage is really what enables you to not have to build the coal plant across the street from the solar plant, right? Like what they do now is you build a big solar field and then you put a natural gas plant across a street to back it up. And energy storage has the ability to take the place of that backup gas plant. We have plenty of excess capacity of like power plants that sit idle and don’t output their full capacity. We’re not lacking in total energy. We have a time-shifting problem where the energy we want is not available when we need it. And that is cheapest and most effective to solve with storing what you have rather than just constantly building.

You know, I told someone an analogy. You know, it’s like imagine you have a grocery store that supplies food for a caterer. And the caterer is expected to deliver their food at the event on time. If the caterer is late, you don’t go and then say, you know, we need to build more grocery stores. The caterer actually just needs a place to store the food so that they can actually deliver on time and make sure that they can buffer the food, right? And so that’s kind of the problem we have. And right now, it’s kind when you have a hammer, everything’s a nail. The commercialized technology we have available to us is to build more gas plants. And so, that’s what we’re doing. And we’re trying to change that by offering storage as an alternative that’s cheaper and faster.

RS: What sort of time frame are you working on at Fourth Power?

AH: Yeah, given where we are today, I would say within five years you’ll see first commercial operating systems. So, we’re targeting the 2028 to 2030 timeframe for first wave of initial deployments. So, we have the first system, integrated system that we’re looking to build next year in 2026. After that is then a pilot system that will be fielded outside working directly with a utility that is effectively 10 times larger, so about 10-megawatt hours in size. And then the system right after that and call it the 2028 to 2030 timeframe would be a commercial operating system that is another 25 times larger in terms of power and energy.

KM: I just wonder when you think about putting on your hat of realism and thinking about storage, which is something you care a lot about and are working on, play it out for us for the next five years. What do you think is going to happen when it comes to storage? And as you said, like, that’s just such a seminal thing and thinking about transition and it’s such an important problem to solve.

AH: This is something that I think I have a different opinion on than many folks, which is, I think there’s a natural tendency when you look at other areas of technology to assume that there is going to be this plethora of options of different types of technologies that will all have their special role to play and do different things. I could be wrong. It may turn out to be that way, but I think the problem of energy storage is so kind of straightforward in terms of what value needs to be given to the grid, that you may see a single or maybe two technologies kind of win out. And the corollary I have for this, or the analogy I have for this, is just look at the existing power infrastructure. There is a single technology we use for transmitting electricity. There is a single, there’s two different types of power plants we build. There’s the Rankine Cycle and the  . They won out. They beat the Otto Cycle, they beat like everything else. It’s a monopoly. There’s like a handful, now there’s many suppliers. You can have a diversity of different, I’m not saying there’s a single company that will win, but there likely will be a single or a very small handful set of approaches that end up really dominating because they meet the needs the best and many other things will fizzle out. And so I think I put so much time and energy into Fourth Power because I think it’s going to be one of these key winners that ends up out there in the end.

KM: And it makes so much sense to me because also, you need to be able to train people to build a certain way. And if you’re like, well, there’s six different things on the menu, that’s a complicated menu. But you know, if you just say, like, we either do it this way, or this way, that’s a simple menu, you can train people. It’s easy to replicate, I feel like.

AH: Yeah, there’s so many aspects of like, consumer technologies and the expectations around consumer technologies that do not port over to energy technologies. You know, so like, for example, I’ll just stick with cell phones, right? Like, you know, Apple invented this multi touchscreen, and it pretty much took over and you got a variety of different flavors, right. And it’s being used on iPads and all kinds of other stuff. You know when you look at, you know as I mentioned, power plants, there’s like basically two inventions. If you look at cars, we all use the auto cycle. You know what I mean? There’s like a bunch of other thermodynamic cycles for like making engines and we don’t use any of them. If you look at heating and cooling and refrigeration, right, we use the vapor compression cycle. It’s like over a hundred years old and it’s still the best. It’s still thing we all use. It’s a single technology that’s winning. I mean, there’ve been a number of other technologies that have been proposed. It can’t compete.

RS: So incumbency is always a problem and it’s part of what you’re talking about here I think. I mean with silicon PV, this is a technology that’s 70 years old, at least in the 60s since it began to be commercial. It’s not the most efficient solar technology you can think of. It’s got all kinds of disadvantages. But one way or another, through either government incentives or just ease of manufacturing and strategies that governments like the Chinese government adopted, for example, of bringing it to scale and subsidizing it, it has reached such an enormous scale of deployment. And along with it, the whole ecosystem of power electronics and battery systems and ways of thinking, And to reach such a giant scale that it’s very hard for better technologies to overtake it. That’s also true, some would say of lithium-ion batteries that have reached this enormous scale now and better alternatives just can’t possibly get there. What do you do about that? I mean to some extent what you’re suggesting is that your technologies are so super fantastic that we should implement them and the government should subsidize them and then they’d become incumbents, and we’d have the same problem. We sort of get in the way of creative destruction that could lead to another generation of technologies.

AH: I think you’re right. And I think that’s the position that myself and many technology developers and inventors we operate from. If you can’t beat the existing state of the art, you don’t get to play. Full stop. And so you’ve got to offer something above and beyond what exists today, otherwise you should go home.

RS: It’s an enormously high standard. It’s not the end point that you have to match. It’s today’s performance and costs that you have to match and it’s incredibly hard.

AH: It is. I mean this is why, you know, like I said very much earlier, right, I really think we should be lauding and celebrating, and in some way, like really commemorating and rewarding everyone and all the various people that worked on bringing wind turbines and PV to the cost that they’re at now. I still remember in 2011, you know, we had a term that we used to throw around, we used called it “grid parody.” When the cost of PV gets to a dollar/a watt and it’s like the same cost as a turbine, it’s a big deal, right? And we got to grid parity and then we passed it. It got even cheaper, right. And we celebrated at that moment. That celebration to me should still continue. It’s such an amazing feat to have brought a technology, like you said, from, you know, a hundred times more expensive than the state-of-the-art to being half the cost of the state-of-the-art and to displace it. That happens so rarely in energy that it definitely needs to be respected and praised.

And we’re trying to do the exact same thing in storage. I mean, storage is a bit different because there isn’t really a widely proliferated state-of-the-art already. The form of energy storage we use today is basically leaving fossil fuels in the ground. We have one-time-use storage. So it’s all stockpiled in the ground. We draw from it. We use it at the rate that we want. But there isn’t really a rechargeable form of storage above ground that has been widely proliferated. Lithium-ion batteries are working their way out, but they’re limited on cost. And so they’re gonna very quickly push up against a ceiling in terms of how much deployment is gonna happen due to cost. And so we are trying to prep for the second wave and come right behind them with something better that’s much cheaper.

KM: I feel like we need to have this conversation in five years to see where things are.

AH: Yeah.

KM: Asegun Henry, thank you so much. This has been great.

AH: Thank you for having me.

RS: Thanks, Asegun.

KM: What if it works? is a production of the MIT Energy Initiative. If you like the show, please leave us a review or invite a friend to listen. And remember to subscribe on Apple Podcasts, Spotify, or wherever you get your podcasts. You can find an archive of every episode, all of our show notes and a lot more at energy.mit.edu/podcasts and you can learn more about the work of the Energy initiative and the energy transition at energy.mit.edu. Our original podcast artwork is by Zeitler Design. Special thanks to all the people at MITEI and MIT who make this show possible. I’m Kara Miller.

RS: And I’m Rob Soner.

KM: Thanks for listening.