What do you do with a 100-hour battery?

Form Energy’s Mateo Jaramillo argues that utilities can use multiday storage to do more than balance intermittent renewables.

December 14, 2023
Listen to the episode on:
Catalyst podcast by Latitude Media
Catalyst podcast by Latitude Media

It’s time to get specific. In the power industry, ​“long-duration energy storage” could mean anything from 4 to 10 to 100 hours of energy. But Form Energy’s Mateo Jaramillo argues that batteries in the ballpark of 100 hours hit a sweet spot, and he thinks that sweet spot deserves its own name: ​“multiday storage.”

In the 15-minute to 12-hour range, lithium-ion batteries shine, effectively displacing the natural-gas peaker plants that run less than 5% of the year. But they don’t displace higher-capacity generation. Nor do they meet the needs of the grid during significant weather events, like heat domes, nor’easters and freak Texas winter storms that can last upward of 75 hours. And for that, Mateo says we need multiday storage.

Form Energy’s iron-air batteries made headlines back in 2021 for promising to deliver tens of hours of storage at a low cost per kilowatt-hour. (Energy Impact Partners, where Shayle is a partner, invests in Form Energy.) So what role could multiday storage play on the grid?

In this episode of Catalyst, Shayle talks to Mateo about real-world examples from Form’s experience with utilities Xcel and Georgia Power. They also cover topics including:

  • The strengths and limitations of lithium-ion batteries on the grid today, and why Mateo thinks lithium-ion is here to stay.
  • The competitive landscape for multiday storage, including iron air, carbon capture and storage, hydrogen and transmission.
  • The roles multiday storage can fulfill for utilities beyond balancing renewables, such as meeting load growth and resilience goals.
  • Bonus: Shayle’s idea for bitcoin mining on a barge.

Recommended resources

  • Canary Media: Form Energy closes its biggest deal yet for long-duration energy storage
  • Carbon Copy: A groundbreaking long-duration battery nears industrial scale
  • Wall Street Journal: Old West Virginia steel mill becomes a green-energy powerhouse
Listen to the episode on:


Shayle Kann: I'm Shayle Kann and this is Catalyst.

Mateo Jaramillo: What having a cost-effective 100-hour battery does for them is it sits as an asset that allows them to bring on these lower cost but intermittent resources, which help them meet load growth while not sacrificing on reliability, or capacity, and keeping cost in line.

Shayle Kann: What would you do with a 100-hour battery? If you don't get the Klondike bar reference then that's on you.

I'm Shayle Kann. I invest in revolutionary climate technologies at Energy Impact Partners. Welcome. Okay, so one of my biggest dumb pet peeves is this class of terms that become popular and then get used and abused so much they basically lose all meaning. And in my mind, there's no better example of this, at least in the energy sector, than the term long duration energy storage. I've heard this term used for everything from six hours of energy storage to seasonal energy storage. I've seen project announcements, and RFPs, where lithium-ion batteries claim to be long duration energy storage despite serving basically the exact same purpose that every other lithium-ion battery on the grid serves. And it's frustrating not just because of the semantics, though I am a stickler for semantics. The basic premise behind long duration energy storage, which is that an increasing share of intermittent generation, largely from wind and solar, is going to command energy storage systems that last longer than the batteries that we've been putting on the grid thus far is true.

Fundamentally, that is a thing that we are likely to need as we decarbonize the grid, but it's not one thing. As the grid evolves, we're going to need multiple types of different resources serving multiple different purposes, some storage, some generation, some transmission, et cetera, et cetera. And one of those new types of resources that presents potentially a new class of asset, comes from Form Energy, which is really the only sizable company that is building a 100-hour battery, or as they call it a multi-day storage asset. So it's a fundamentally different type of asset than certainly the lithium-ion systems that we see on the grid today, but even I think most of the other so-called long duration energy storage systems that are being discussed or installed today. So it takes some wrapping your head around exactly what role it might serve on the grid, what it competes with, how it might fit into the picture of a fully decarbonized grid.

And it's kind of important, I think, because there are a bunch of problems that you need to solve as you increase generation from intermittent resources on the grid. Some of them are hard to solve with the tools that we've got at hand today. We at EIP are investors in Form, and I would say this is the most common question that we get, which is what do you actually do with a 100-hour battery on the grid, and do we need it? So let's find out. Not for the first time and probably not for the last, I brought on Mateo Jaramillo, who is my friend and the CEO and co-founder of Form, to talk about the various classes of energy storage that the future holds, and what one actually does with multiple days of energy storage sitting in a battery. Here's Mateo. Mateo, welcome.

Mateo Jaramillo: Thanks, Shayle. It's great to chat with you as always.

Shayle Kann: As always, excited to chat with you about the future of the grid and the role of multi-day storage, but let's build up to it. Starting with talking about the predominant form of battery energy storage on the grid today, which is lithium-ion. So you were involved in the early days of building Tesla's stationary energy storage business and product before starting Form. So let's talk about the role that lithium-ion plays on the grid, how you saw that progressing then and where you see it today. And then we'll use that to build into, what might we need in additional lithium-ion and why? But what is the role of lithium-ion as you see it?

Mateo Jaramillo: Sure. So as always, again, great to be back. But the role of lithium-ion is becoming pretty clear. It's already, obviously, used at great volumes today, although I do wonder how much lead acid is still deployed on the grid that we have forgotten that's sitting around out there.

Shayle Kann: It's more than you'd think. I actually saw those numbers not too long ago. We're not adding a ton of new lead acid, but there's still quite a bit there.

Mateo Jaramillo: It's a stale capacity for sure. But I did work in lead acid before I worked in lithium-ion, so I do have that for me. But yeah, there's a growing acceptance of the value of lithium-ion to provide relatively short duration functions into the system. And that's for everything from minutes of duration, where it originally started, fifteen-minute fast frequency response in PJM to now longer duration. That's the key, longer. Four hours, coming up on six hours in some cases. And that's really providing a lot of peaking services or ramping services into the system. So helping with the balancing function, broadly speaking.

And although they are capable of providing energy for hours at a time, still they're essentially power batteries, largely. And they're doing intraday applications, whatever they may be. So generally, taking energy from one part of the day and moving it into another part of that same day. And that's sort of the broad umbrella for what lithium-ion is doing on the grid today. And it's doing it for a lot of different regions geographically, for a lot of different specific functions within that, but that's generally what it's doing.

Shayle Kann: And I'm sure a lot of our listeners will already know this, but just as a reminder for anybody who isn't super familiar with how the economics scale for lithium-ion, technically there's no reason lithium-ion couldn't deliver days, weeks, seasons of... There's no technical limitation to the duration of lithium-ion batteries, an economic limitation, right?

Mateo Jaramillo: Yeah, that's absolutely correct. And lithium-ion has fantastic technical capabilities for a lot of reasons, very dense gravimetrically and volumetrically. It cycles a lot. And it can discharge at much slower rates than what is currently used for today, like I said, minutes to hours duration. But it becomes economic to do so over that time period. So that's sort of the key trick, when you talk about duration of a energy storage technology, it could be any kind, you really have to make sure that we're talking about at rated power. So when we rate a battery, we rate it for both energy and power, and the energy is only relevant to the extent that it's in direct connection with the rated power of the system. So you could take a four-hour "rated lithium-ion battery" and discharge it for 100 hours, and you would effectively be paying a much higher price for that power over that duration.

Shayle Kann: And why is it that you wouldn't design or shouldn't design a lithium-ion battery to do dozens, hundreds of hours of storage at rated power?

Mateo Jaramillo: It's really just cost. So you have a per unit of per hour cost, essentially, adder, and for every incremental hour that you want to be able to discharge at rated power, you pay that cost. In the case of lithium-ion, let's say that's roughly $100 a kilowatt-hour, you're adding $100 for every hour you want to discharge. And the industry normalizes around dollar per kilowatt in terms of comparative resources. So everything comes back to that figure in the end. In the battery world, we like to talk about kilowatt-hours, dollar cost per kilowatt-hour, but the industry does not really do that. And so your cost comparisons are always on a dollar per kilowatt. And so then the question becomes, how many hours do I get for that cost per kilowatt? And for lithium-ion today, that's roughly four hours. But if I need 100 hours, I'm not going to pay 25 times as much on a per kilowatt basis for that resource. It just doesn't clear in the market. It is not anywhere close to being a least cost alternative to provide that kind of function into the system.

Shayle Kann: Right. Okay. So we'll move on from lithium-ion to talking about what technologies can serve the longer duration applications and what those applications actually are in a second. But before we do, one thing I'm curious about your perspective on is, do you see any... Is lithium-ion just going to remain in perpetuity dominant in that category, in that 15 minute to four hour duration category? Do you see any reason why a different technology would play a significant role or do you think that at this point the train has left the station?

Mateo Jaramillo: Well, it certainly is the incumbent, and displacing the incumbent in this industry in particular is challenging. I also think that there's a cost entitlement to be significantly lower than where it is today, which is low cost compared to where it was just a few years ago. So it's a shifting landscape for any sort of aspirant to dethrone lithium-ion in that, I would say, intraday duration space. So ultimately, that's roughly 10 to 12 hours for a complete cycle, let's say, in the use case that you're going for. And there may be others that come along or are a bit cheaper, but again, displacing lithium-ion is going to be really hard.

The grid markets are sort of the tail on the dog of the automotive markets right now, driving the production of lithium-ion at scale. And so the industry that we're in, the electric industry, gets to benefit from that massive scale, and there is no other chemistry or technology that benefits from the same tailwinds, essentially. So it's really hard to see how that happens precisely. I'm a technologist optimist at heart, so I like to believe that new great technologies will show up in some interesting ways, but I also think it's a really tall task to think that brand new technology from a standing stop, essentially, today, is going to be displacing lithium-ion across the board for anything up to 10 or 12 hours.

Shayle Kann: Okay, so let's move on to talking about the "long duration energy storage world," which I think you agree with me is a term that gets abused more than it gets appropriately used. So let's start by talking about what we... Generally, people mean durations longer than... I think people mean. Actually, I don't know what people really mean when they use the term, and I'm sure it differs, but in theory it's longer durations than what today's batteries typically deliver. And so I've heard everything from six hours to seasons being within the long duration energy storage category.

And one of the things that you and I, I know, have talked about, is part of the result of that is that there've been some solicitations from utilities that are for "long duration storage" that'll be for eight hours plus or something like that, and lithium-ion just ends up winning that still today. So let's not talk about long-duration energy storage as one category, let's talk about it based on the actual functions of the batteries that are being deployed. Beyond the, as you said, the four to let's say eight-hour range that lithium-ion is well suited to, do you think of there being value in a middle category before we get to what you're up to that's in that, I don't know, 12 to 24 hour range?

Mateo Jaramillo: So let me paint the landscape a little bit more broadly because I think it is helpful. We at Form anyway always think about the system operating as a portfolio. And within that operation of the portfolio, what I mean is all the different assets, whether they're generation assets, or transmission, or storage, they all need to, obviously, operate in concert. They need to be synchronized. Literally for frequency reasons and then also just in general like operating a fleet. And it's really important within that context when we're thinking about the function of duration in there, for storage in particular, to understand how that duration fits into, again, that portfolio operation. What is it doing? What kind of value does it bring through those numbers of hours that we're saying we have? And I think what lithium-ion has done, as we've talked about, has already established very clearly the value of a few hours of duration.

Right now it's predominantly four hours. Let's say it's moving to six hours in the applications that I mentioned earlier. So there's unambiguous value for that duration of energy storage. And then we start to say, okay, well, what additional value is there through the incremental hours of duration? And that could be 12 hours, or that could be 24 hours, could be 45 hours. And so then we need to really understand what kinds of things it might displace. Let's say you built a model and you ran the co-optimization, what would, let's say lower cost lithium-ion or something like it, what would it displace in the system? If I run my capacity expansion models, I run my integrated resource planning, what gets picked up instead of, or rather, what does picking up that duration energy storage displace that I otherwise would have in my system? And one way to think about it is sort of capacity factors for gas plants.

And so we say peaker plants, that's sort of the easy way to say that's what lithium-ion does today. It displaces peaker plants, gas peaker plants. And peaker plants typically are defined as gas plants that operate less than 5% of the hours in a year. And so if you take the total number of hours in the year, that's 400 hours roughly. Lithium-ion batteries can relatively easily accomplish that kind of thing. By the way, these are not consecutive hours, these are spread out over the course of the year, so that's why that works. And then you start to go through the histogram of the capacity factors of the plants that are operating on the grid, and these fall into a big bucket as what the industry would call mid-merit gas plants. So 20% capacity factors, 30%, all the way up to 60 or 70%. And then there are some plants, of course, that want to be operating very close to 100%, and those are what we historically would call base load plants.

Base load is a notion that is sort of going away for lots of reasons that we probably don't want to go into right now. But that question of duration for energy storage needs to be answered fairly precisely in terms of function that those other mid-merit resources are providing today. And so what we find in the modeling is that there's a lot of value for up to 10 or 12 hours, let's say. There is less value. I'm not going to put a specific number, but there is less value that we see very markedly in the system until you get back up to about 75, coming in on 100 hours. And that's because with that duration, once you're close to 100 hours, you can functionally start to replace what those mid-merit gas plants are doing 30, 50, 70% gas plants.

And so that was sort of one of the key insights from the work that we did originally at Form on the analytics side of things led by Marco Ferrara, my co-founder, that really drove us to that duration. It's a lot murkier the precise value you get going from, let's say 10 to 20 hours, or 20 hours to 40 hours. You can't do much more with that incremental duration that you have to pay for, of course, in the device, that somebody's going to compensate you for. The utility, the market, whomever. And so I would say that that still needs to be worked out. There are some cases you could maybe look at day to day shifting, but the values there, again, they are not nearly as clear as the value you can bring to the system once you have 100-hour duration, four days of duration.

Shayle Kann: Right. So let's get to it then. So Form is building a... First product is a 100-hour rated battery. Let's talk about why that. You sort of alluded to it, you can start to displace some of these mid-merit gas plants. But as you said, these are operating at 30%, let's say, of the hours out of the year. But the number of consecutive hours that they're operating is sort of equally important in this context, right?

Mateo Jaramillo: Yeah, that's right. And it's a combination of total hours, as well as consecutive hours when needed. These are, in the end, playing a role of capacity, reliability, on the system, and so it is really important to zoom in to very high fidelity to take a look at what precisely it's doing when. And the reason why that number of hours, four days roughly, becomes relevant is because that's where a lot of the material weather events sit in terms of duration. And in a grid that increasingly is driven by weather, renewable energy, you need to be able to account for that type of intermittency that does inevitably come along with weather, regardless of what region you're in geographically. It could be a heat dome in the Pacific Northwest, or a polar vortex in the Upper Midwest, or a nor'easter in the Northeast, or an unlikely Winter Storm Uri in Texas. So you really need to be able to account for that duration because the signature of volatile weather events is frequently around that duration, three, four, or five days.

Shayle Kann: All right, I want to give some more specific examples because one of the things that I've heard a fair bit, I suspect you have as well, is this idea that very long duration energy storage, multi-day storage, whatever we want to call it, is probably a necessary component of some future, extremely high penetration renewables world, but certainly not necessary and perhaps not even that valuable today. We don't need it yet, that's basically the argument. I don't think that's true, but I think it's because people don't fully understand the use case. So Form has announced a bunch of specific deals with specific customers, maybe we could talk through some of them and use them as archetypes for where there's value today in a 100-hour battery, and maybe how that'll change in the future as we do get increasing penetration of renewables. But pick your starting place, give me an archetype.

Mateo Jaramillo: Yeah. And I think it's really important as well to locate the duration in the cost consideration as well. It's always 100 hours, but at what cost, right? And I think that the assumption of, oh, that duration is not relevant, makes a really big assumption about how much you would have to pay to get it. And so what we have always targeted at Form was, it is a cost to the customer where it is unambiguously valuable to have those 100-hour duration... We are today paying for 100-hour duration resources, they just don't happen to be storage and they are compensated in the original, let's say, construction of the market design. In other words, we don't really pay for reliability, we don't precisely pay for reliability or capacity. Sometimes I use those terms interchangeably, although there are more precise definitions than that.

But for the purpose of this conversation, that is a way, probably a good way, to think about what this type of resource is doing in here. And so when we target 100 hours, it is at a pretty precise target cost. And coming out of the modeling that we did and what gave us the courage and the confidence to go build the company, all the modeling that we did confirmed by third-party modeling is that 100 hours and $20 per kilowatt-hour is sort of the magic combination. And if you can hit $20 per kilowatt-hour, the system will pay for 100 hours of duration. And the reason it will want to pay for 100-hour duration is because you're solving problems that you can only solve with that duration, but you must have that cost to go after it. And so that's where we see that showing up in the system today, is you're able to achieve whatever goals that you may have as a system operator, it could be utility, let's say, or an entire market.

Whatever goals you have, if you have cost-effective multi-day storage, it just makes meeting those goals easier. That's one way to think about it. It's a new asset. And right now we are under-optimized for types of storage assets to play on the grid. Right now we only have short duration, whether lithium-ion or anything else that is cost-effective for that application. What we do not have today is cost-effective multi-day storage. But what we see, again, over and over again in the modeling, once you introduce that cost-effective multi-day duration asset into the mix, it makes whatever your goals are, again, that you need to achieve, they could be reliability goals, it could be capacity expansion goals, in other words, meeting load growth goals. It could be meeting decarbonization goals. Resilience, it doesn't matter. If you have this new type of asset, it just makes it easier.

So that is why we are very confident that the market is ready for this, we just need to hit the cost points. And then it becomes a question of, okay, how does Form hit those cost targets? There's a whole bunch of reasons why we're confident about that, but we're well on the path to being able to introduce, at scale, that cost-effective multi-day storage asset. It's an asset class in many ways that we're trying to create more than a battery per se. And what you're getting at with these customer commitments is that's what they see. That's why they're buying in now. It's because they see that the targets that we're putting in front of them, which are credible for the scaled resource, enable them to bring more value into their system and deliver a better product, i.e. reliable, cost-effective, decarbonized electricity to their customers easier.

Shayle Kann: Okay, so theory of the case is that this multi-day storage asset class can help utilities achieve a bunch of these various goals that they have along the way to decarbonization. Before we move on to some specific examples of that, I do want to talk about that $20 per kilowatt-hour cost figure that you cited because contextually lithium-ion cells might be $100 per kilowatt-hour. Lithium-ion systems fully installed on the grid, two, three times that. So we're talking about an order of magnitude reduction from a dollar per kilowatt-hour cost perspective relative to lithium-ion on the grid today. Let's talk about cost entitlement. What makes it possible that Form could get to $20 a kilowatt-hour fully installed with an iron-air battery?

Mateo Jaramillo: Yeah, starting from what kind of chemistry, what kind of embodiment of that chemistry would even come close. That has an entitlement to be less than $20 a kilowatt-hour. And so we really started from the fundamentals, and iron-air is what we ended up with. We did not invent the chemistry, it's been around for some time, but it's never been commercialized. But what we saw was the entitlement is there, cost of materials, the mechanical designs that embody that, the O&M costs, the cost to manufacture, you name it, the raw abundance. It all is able to scale and is able to hit those cost targets.

One quick point of reference, lithium-ion active materials. So you pull the active materials out of the ground, you put them on a table, it's maybe 30, $35 a kilowatt-hour, unprocessed, not turned into a cell at all. For us, it's less than a dollar per kilowatt-hour. And so you start with something very, very cheap, and the trick, of course, is to end up with something that is also very, very cheap. I can't do any expensive synthesis and I can't do any fancy manufacturing processes, high precision type things, so it's got to stay cheap. And that's the real trick that the company has really innovated in, is how do you start with something that is fundamentally cheap and end up with a device that, in the end, is a piece of infrastructure that remains a very, very low cost?

Shayle Kann: All right, so let's get into some real world examples here. Can you talk about, maybe just pick your favorite one, what's an example of how an iron-air battery, how multi-day storage is helping a utility achieve their goals around whether it be load growth, decarbonization, reliability, et cetera?

Mateo Jaramillo: So take Georgia Power. Georgia Power is the main utility in the state of Georgia, no surprise. And we're doing a project with them. It's slated to be about 15 megawatts, so on a power basis, not so large, but about 1,500 megawatt hours.

Shayle Kann: I just want to pause on that for one second because 15 megawatts, if you're in the electricity industry you know that sounds kind of small, but the funny math that you have to remind yourself of when you're doing 100-hour battery is that it would be a 1,500 megawatt hour battery, which as far as I know, is going to be one of the three or four largest batteries by energy capacity in the world.

Mateo Jaramillo: Yep. Assuming nothing huge shows up online in the next couple of years, which it might. But yes, it's a large energy basis for that battery. And Georgia Power is a very, or the state of Georgia, is an interesting grid right now. It's quite dynamic. There's a lot of load growth, there's a lot of population growth, a lot of industry growth in the state of Georgia. And in fact, the load growth is coming on so quickly that Georgia Power has gone back to their regular to update their integrated resource plan two years early because they see the need to build out much more capacity even sooner than what they had anticipated. And they want to be able to do a lot more renewable power than what they are currently on track to do today. And that's based on customer demand, large industrial customers who want there to be clean energy in Georgia, as well as it being a very low cost resource that they want to be able to incorporate into their system.

And so part of what having a cost-effective 100-hour battery does for them, is it sits as an asset that allows them to bring on these lower cost but intermittent resources, which help them meet load growth, while not sacrificing on reliability or capacity, and keeping cost in line. So it sits across a couple of interesting different value streams for the utility. And what's great about Georgia Power is they, at least in this case, they run their own internal markets. They're not part of a wholesale market, and so they can pretty precisely put a value on reliability, which the wholesale markets today do not do, and so they know how much they're willing to essentially pay for reliability. Doesn't mean they tell us, it just means that they know what their willingness is internally, and then, of course, we have to negotiate it and it has to be approved by their regulator and everything else. But that's an example where this kind of asset just makes them meeting all those goals that they've got for load growth, and decarbonization, and reliability, it just makes it easier.

Shayle Kann: The other factor too, I think, if you're in one of these situations where you're building out a ton of new generation, which is definitely true of Georgia Power, as you said, is that the introduction of multi-day storage actually affects the amount of new generation that you need to build, right?

Mateo Jaramillo: Yeah, that's right. And these days the plans are to build so much new generation overall, and so much new renewable generation specifically. And even within that, so much solar specifically that the implication on the land is quite large in fact. And if you take a state like Georgia or the plans that they have, it's implying quite a bit of land. Now, to put it in perspective because I don't want to be a land doomer on this, it is a tiny percentage of land that's put for all sorts of industrial uses, so it's not like you're blanketing the state. However, going and getting land for any power project, frankly, is challenging these days. And so by having cost-effective multi-day storage, you can incorporate more renewable energy without sacrificing reliability at basically half the land cost.

And by that I mean you only have to put solar on half as many acres to get the same benefit. And that's because you're essentially solving the reliability with a different type of asset that allows you to co-optimize your system in a different way. And this is a big driver. There are very public battles over land use these days for power projects pretty much of any kind, but in particular for places where they want to build out a lot of solar, so that is another benefit. And we see this, again, over and over, it doesn't matter whether you're trying to build a lot more wind resources, offshore, or onshore, or solar. Having this kind of asset makes, again, achieving your goals just a lot easier in this particular case by really reducing the amount of land you need.

Shayle Kann: All right, I want to do maybe two more examples. So we've got Georgia Power example is Southeast, US, load growth, lots of new renewables, solar heavy part of the country, let's try the Midwest. Wind heavy as opposed to solar heavy. Different weather. We didn't really talk about the weather issues that Georgia Power faces, but they're different if you're in MISO. And organized wholesale market, which is not true there.

Mateo Jaramillo: So let's take one of the Xcel Energy projects. Xcel is one of the major utilities in the Midwest in the United States. They're based in Colorado, but they've got operations south of there, all the way up to and including Minnesota. And they are a wind heavy mix that they have there. And they also participate, as you were saying, in wholesale markets, in this case MISO. And so for them, they want to be able to hit their decarbonization goals, and Xcel Energy deserves a lot of credit for being the first major utility that really put a public stake out there and said, "We intend to be 80% decarbonized by 2030, and we have the tools to go do that right now. We intend to be at 100% by..." I think they said 2045. "We do not know how we're going to get there, but we're still going to make the commitment and we assume that the technologies that allow us to do that cost-effectively will show up."

So they really put a huge stake in the ground and at least informs little tiny arc. That was a big moment because we were still fairly nascent. We are still. But even more nascent company. And that was a big validation that, contrary to your earlier point, there was going to be a need for 100-hour, longer duration, let's say, sooner than what people were anticipating and Xcel sort of put that goal out there. So we're working with them. In this case, it is still around 100-hour duration, which allows them to incorporate more renewables into their system while meeting their load growth. Some of the dynamics are all very similar. We are in a growth moment for the electric industry, which is not what it had been for the last 40 years roughly. And so all the utilities are thinking about that and they're thinking about how they meet load growth while they have assets which have recently retired, assets which will soon retire, specifically coal in this case.

And also how they, again, add as much new generation as possible. And so Xcel also participates in that wholesale market you were referencing, Shayle, the mid-continent system there. And so for them, there's a market element to what they want to do, they want to be able to bid their assets in right ways and hedge their risk, their cost essentially, for being exposed on the wholesale markets too. And that's what this type of asset allows them to do. Having 100-hour duration resource is a physical hedge against price spikes in a way that you can only really financially hedge right now unless you have, again, thermal resources. So it sort of opens up a new frontier for market participation in a way that short-duration storage does not.

Shayle Kann: All right, one more example I want to talk about. Maybe this isn't a specific deal that you've announced, but because it's, I think, sort of the topic du jour right now in the industry, which is the transmission and interconnection bottleneck. As you said, we're in a resumed period of growth in electricity demand, but there is a ton of generation that's sitting in interconnection queues, there's reform coming, but who knows how big it's going to be and how soon it's going to come. And meanwhile, we're really just not building out enough new transmission. I think everybody now recognizes this is a big problem and becoming bigger as time goes on. In the early days of Form when you were still figuring out, what is the type of battery we need to build and what role might it play? One of the proposed use cases was alleviate transmission bottlenecks. How do you see that playing out today?

Mateo Jaramillo: Well, it's moving quickly, I would say. It was a application that we identified early on and then we put on the back burner as we went and pursued the ones that we just talked about. But it's come back pretty quickly, partially because people have really picked up their wind plans again in the wake of the IRA getting passed. And what we're seeing now is there's that application in a couple ways. One is just sort of allowing more capacity on the existing system as is. So you have congestion, you have curtailment as a result of a lot of wind showing up at the same time on the same node. And so how do you deal with that as a transmission system? And so we look at it from that view. How do we build out more transmission as efficiently as possible? It's difficult to build transmission even when you have great people like Mike Skelly working on that. So we want to be as impactful for every line that we run for transmission. And having storage, the ability to buffer at one end or both ends, makes that even more a highly utilized resource.

So that's one angle to it. The other is on the operational side of things, if you own one of those wind farms and you're now being curtailed, or let's say you're in the market and you have a ton of basis risk. In other words, my cost that I'm paid is at a different location from the place where I'm putting in my power and I bear that risk for that price spread. The ability to have a type of resource that allows you to, what we call shape around that risk. In other words, move your energy by days, essentially, is what's required to get out of that conundrum that a lot of existing plants find themselves in today. And so that's sort of a, I would say, fast coming application for us. We're looking at projects with project owners all throughout SPP, the Southern Power Pool, all throughout MISO where there's a lot of wind today, and a lot of existing assets that are in some challenged economic circumstances. And again, having this new physical hedge helps maybe turn around some of those assets.

Shayle Kann: The other thing I've heard, I think, a number of times when I talk to folks about Form is the notion of what do you do... Okay, so let's just say you've got a 100-hour battery on the grid, and this is going to be specific to the application and the situation, but let's try to generalize it if we can. What do you do with that battery? Because the simplified version of it is, if it's a capacity or reliability resource predominantly, and if it looks like a gas peaker does or something like that.

Or similarly, if you're operating in ERCOT and you figure you're going to make all your money off of the three days a year where there's some multi-day crazy weather event and prices are $9,000 a megawatt hour, then you could imagine the operations of the battery are super, super simple. You basically charge it up and then just hold the charge and wait for your window, and then discharge for three days straight, and then charge it back up until the next one. Is that how you see operations of a Form battery in most situations or do you think it's going to be more dynamic? What will it actually look like?

Mateo Jaramillo: Yeah, I think it will not be that first case at all. I think it'll be much more the latter. That arbitrage play that you're describing in ERCOT, that is what lithium-ion does today, and it can clear the return hurdles for a merchant plant in that market based on the structure of that market. Right now there aren't enough hours that have a long enough spread where it makes sense to have a 100-hour battery doing that. It does make sense for a four hour or two hour battery, or even a one-hour battery, frankly, in ERCOT. So that is not the main application that we're going for. We're not trying to arbitrage an energy only market and moving around those kilowatt-hours or megawatt-hours. Rather, what we find, again, in all the cases, somewhat unique but common, is that the bulk of the value is for that capacity reliability resource, and that's how you're justifying the cost of it.

And you will use it as much as you possibly can. It's sort of like every other resource on the grid, it's like utilities go to their commissions and they get assets approved for the original case that they had in mind. And then once it's a asset on the grid, you're going to use it as much as you possibly can to the highest value. You're never going to use it to negative value, but you'll use it for a lot of incremental value. And if we look at the dispatch profile for this kind of battery over the course of a year, and we look at the state of charge over the course of a year, we find that a few times per year, inevitably, and when this happens, depends on what kind of grid you're in, but inevitably you're discharging flat out for three or four days because of a weather event.

But we also find that there's a lot of activity that's happening intra-season. So you may have a oversupply of renewables in the spring and demand is relatively low, and so you're sort of ratcheting up the state of charge over the course of two or three weeks because that's the right thing to do. And then in summer, you may have a bit of a deficit that you accrue over the course of a month, let's say, or six weeks. And in that case, you're ratcheting down the state of charge for the batteries, you're net discharging into the system. Or it may be doing ramping support, if that's a valuable thing to do. So when we actually dispatch it per the profile optimization of a given utility, in the end it's doing a lot of things. Its core reason for being, of course, is that multi-day duration capacity resource. And then on top of it, you're using it for a lot of things.

Shayle Kann: Yeah, and I've seen these profiles that you guys have put together, hourly profiles of what it's going to look like. It's lots of shallow cycling up and down, but with a clear net trend, and that net trend lasts days to weeks to seasons. But in between during that time it's a net trend, so there's lots of ups and downs in the meantime because you're charging and discharging to do whatever other purpose you have.

Mateo Jaramillo: That's right. And I should be clear, 100 hours, nobody at Form believes that 100 hours, a nice round number that suits our human sensibilities, that that's the universe deciding that's the optimum number. And there are some grids that would prefer probably 150 hours or 175 hours, and other grids that may prefer 75 or 80 hours. But what we are doing is introducing a product at 100 hours that allows us to solve the bulk of the market. Down the line, we almost certainly will be optimizing for more niche markets. What does the Irish battery, what should that look like over the longer term, right? Lot of offshore wind, a lot of congestion on the island. Okay, maybe that wants a little bit longer battery. Or Arizona prefers a shorter duration battery because they've got great solar resource year round. So I also want to be clear, it's not like I believe or the modeling would say that you only need exactly one duration and that duration is 100 hours, but it does allow us to address the very large bulk of the market.

Shayle Kann: So I guess stepping back to what this tells us about the future of a decarbonized grid, as you said, it's a mix of resources that we're going to need. I think we agree that a fundamental principle underlying Form's existence, as I think about it, is continued growth of intermittent renewables. That's a fundamental tenet, so let's posit that's going to continue to happen. There's going to be other stuff as well in the decarbonized grid, and so as you think about Form's competitive landscape, I guess, it's not really lithium-ion batteries, it's other things that can serve a capacity reliability type service in a decarbonized fashion. So what's on that list for you?

Mateo Jaramillo: Yeah, we think about it from a substitution standpoint. So what type of resource could perform the same sorts of function? Carbon capture would be one. If you could neatly put a box on top of the flue gas and capture the carbon, that would essentially solve the big challenge there. Now, it would add cost, so again, it all comes down to cost for the industry and we have to be really mindful of that. But that would be an option potentially and there are some pathways that are being explored there. Another would be hydrogen. There's a lot of discussion around hydrogen in the system today. Again, big questions on cost and big questions on scalability of that. At the other end of the spectrum might be more transmission, like a lot more. So if you could wheel a kilowatt-hour from Arizona into New York frictionlessly with no loss anytime of the year, well, that would be beneficial for the system and maybe you don't need as much storage. Also, has challenges for timeline and cost. So those are a couple of things.

Part of what we're going for is speed, however, speed and scale. And one nice thing about batteries is that they can be deployed at scale. They're not terribly complex. We build modular systems, our iron-air battery, it's a meter scale device that we will go build very many of in a factory. And that means we can have high quality, we can have good performance, and we can have good scalability, we can drive our cost down. That's sort of key to everything there. And so we see this as a bit of a race. There are other technologies, hopefully, that do show up on the system that help the electric grid overall, not just meet its decarbonization goals, but also meet its load growth goals. And I would say that's one big difference in the mindset, Shayle, from when we started the company. We sort of were anticipating a huge amount of renewables growth, but I think the overall demand growth is really new to the picture and it's trending that we would double...

We would need a grid, which is twice the size that it is today, by about 2045, 2050 roughly, given the growth rates. And that's not even really digesting the impact for whatever they may be on the large language model compute side of things, which is driving demand bonkers right now. So it's really load growth mixed with more renewables, but it's really every type of resource, generation resource, showing up, trying to help, more or less. We're going to need to build as much solar as we possibly can. We're need to build as much nuclear as we possibly can. We're going to need to build as much wind, and geothermal, and on down the line. Absolutely everything. And so that's sort of the broad contour of the market that we see. And having this kind of asset, cost-effective multi-day storage, just slots right in there and I think can be a very impactful asset class for the entire electric system.

Shayle Kann: I noticed that you did not mention my favorite asset, which is my Bitcoin mine on a barge idea. It spends six months in the Northern Hemisphere, six months in the Southern Hemisphere and just soaks up excess load, or sorry, excess generation in each region. You should just add that to your-

Mateo Jaramillo: We'll do some iron reduction on that barge too. We'll move iron around the world.

Shayle Kann: I guess that's more valuable. Fine. Probably a Web3 skeptic too. Anyway, Mateo, this was fun as always. Thank you so much for taking the time.

Mateo Jaramillo: Thanks, Shayle.

Shayle Kann: Mateo Jaramillo is the co-founder and CEO of Form Energy. This show is a co-production of Latitude Media and Canary Media. You can head over to canarymedia.com for links to today's topics. Latitude Media is supported by Prelude Ventures. Prelude backs visionaries accelerating climate innovation that will reshape the global economy for the betterment of people and planet. Learn more about their portfolio and investment strategy at preludeventures.com. This episode was produced by Daniel Waldorf. Mixing by Roy Campanella and Sean Marquand. Theme song by Sean Marquand. I'm Shayle Kann and this is Catalyst.

No items found.
No items found.
No items found.
No items found.