The Exchanges

Q that most listeners to this podcast are somewhat familiar with the hydrogen world, but probably not so familiar unless they have dedicated time to methane pyrolysis with carbon black. Uh, and Jigar, I know this is a soapbox You want to jump on as well. Tell a little bit more about what do we use carbon black for today? And what is the incumbent production method that this is displacing?

A Yeah, I mean, you know, I don't know that I was the world's expert in carbon black either when I came into the job here at the loan programs office, but, you know, having gone deep down the rabbit hole, I mean, 60% of carbon black is, uh, used in tires, the other 40% is used in plastics and, you know, mechanical rubber, you know, sort of goods. Um, but the way that you produce it now is you basically partially combust, uh, You know, things like coal or, or tar, basically. And, you know, you sort of capture that soot, and that's carbon black, right? And there's 15 major plants in the United States. All of them are under EPA, Department of Justice consent decrees. As of 2013, none of them had had Sox and Nox scrubbers on them, so the people who lived in those communities nearby were breathing in some of the most polluted air in the country, mostly in Louisiana and Texas. And, um, you know, I think some of the carbon black producers are actually still weighing whether they want to comply with those requirements or whether they're just going to shut down their facilities because it's too expensive to put in scrubbers. Um, and so, but they're essential, right? I mean, even an electric car, even in a Tesla, you got tires. And so, you know, like by driving a car, you're participating in, you know, one of these remarkably dirty industrial processes.

AI assessment note: “60% of carbon black is, uh, used in tires, the other 40%”

Q California, like a cratering. Um, but to your point, it cratered, and the remaining market has a very high Battery attach rate. So, you could make an argument that's exactly what the point of the policy was supposed to be. If you believe the value is in the battery attached to the solar, then you need to, like, force a high attach rate, and this is what did that, right?

A Yeah, there's a couple pieces to this. One is that, um, I think we've always known that net metering has declining value beyond sort of five percent penetration, and, but when you get to five percent penetration, the solar industry has so much political power that they're able to keep things going. For a little longer, and so that's sort of what happened in California, and you see that same thing playing out in Puerto Rico right now, where, you know, they passed another law, and so then, when it really gets to the point where you have this cost shift, then, you know, it, it feels abrupt, even though they probably should have been eased in three years earlier. On the battery attachment rate, I totally agree with you, and, you know, but the thing that frustrates me about California, right, is that, They did the NEMM three point O piece, but there's no transparent way of getting paid for your battery piece. So like they didn't do, they did the peanut butter, but they didn't do the chocolate.

AI assessment note: “On the battery attachment rate, I totally agree with you”

Q and then repowered, and as I understand it, at least from the, the articles that have been written about this ACES project, the way they're thinking about doing it is that repowering it on day one with a blend of natural gas and hydrogen, with the hydrogen share increasing steadily over time until ultimately by twenty-forty-five or something like that, it becomes a hundred percent hydrogen Generation. Is that right?

A Yeah, that's right, is that there's an 840 megawatt natural gas plant. Our money is not being used at all for that. So our money, the project that we're funding is an electrolyzer project using a salt dome as storage, right? So none of our money is going to a new natural gas facility. Um, that natural gas facility, uh, is supposed to be using an initial blend of 30%. And then ramping up to a hundred percent, which they think they could do as soon as 2030. Um, but I think that as, as you and I look at the marketplace, right, the role of peaker plants is, uh, going to be moved, uh, increasingly to emergency situations, right? That you've got battery storage and lots of other technologies that are being used inside, uh, the Intermountain West. To be able to provide these sort of services. And so peaker plants, uh, you know, in the, in the new modern grid should be used less and less, uh, you know, during wildfire season or polar vortexes or, or things that may occur. Right. And so, so, so my, my sense is that that hydrogen then could be repurposed at that point to higher value, higher value things.

AI assessment note: “Yeah, that's right, is that there's an 840 megawatt natural gas plant.”

Q for folks who are thinking about, like, what would it take for me to have a project that is attractive? Um, and this is, you know, this is project capital, right? You can also provide manufacturing Capital, but let's set that aside for projects. What are the sort of fundamental characteristics that the ACES project has that are like, okay, checks all the boxes for us. We can move forward.

A Well, The first thing we have is an existing interconnection point, right? Which is hugely important. You can imagine that those are in very short supply today. You also have, you know, access to water because of the existing water rights that, that were held by the coal plant, right? But you also have a group, um, who's actually willing to do a tolling arrangement, right? I mean, when you think about, um, what's the, you know, the structure of the project, um, both the water and And the power is being, um, told by Intermountain Power Agency, right? And so, so, you know, from a risk standpoint, um, you don't have to take the risk on power prices, um, water availability, and all the other pieces in this project, right? So the de-risking thing here is quite substantial. But the other thing I'd say is that as we move forward, right, This project is really well structured. It was rated at investment grade, um, for the loan, so it's actually really well structured. Many of our projects are triple C or single B in terms of its shadow credit rating, and so, um, but once the technology part of this becomes demystified, then you can imagine there's a lot of wind and solar developers who are saying, um, We are frustrated by the fact that we have to accept a below-market PPA price. Remember, we've had this conversation around whether PPA prices were heading towards zero on utility-scale s…

AI assessment note: “The first thing we have is an existing interconnection point, right? Which is hugely important.”

Q uh, you published a report around the end of the year. There, you have 66, this is as of maybe December, so could be different now, but you have 66 active applications for Fifty-four billion dollars in loan requests. It's everything from, like, nuclear to biofuels to transmission, all sorts of different things. Like, what, what should we be expecting to come out of the loan program's office post Monolith?

A Yeah, I mean, you know, I think we've got, um, projects that are across Title XVII, right, which is our sort of project finance type vehicle, and that's where Monolith got its money out of the fossil title. Um, I think you've got, uh, ATVM, right, was the Advanced Technology Vehicle Manufacturing Program. That's electric vehicles, battery gigafactories, critical minerals. Um, and we've got the Tribal Energy Loan Guarantee Program. I think that you're going to see an even split, um, Between the advanced technology vehicle manufacturing program and, uh, the title 17 program in terms of, uh, loans that could come up, that come out of the office. And so I think you're going to see a lot of, you know, EV manufacturing facilities, battery gigafactories, critical minerals, as well as renewable energy, nuclear, and fossil title, you know, project finance deals, um, which also include like fleet deployments, for instance, like fleet deployments of EVs or clean vehicles would be in title 17. And I think you'll see our tribal energy Loan guarantees. But I think, I think the bigger thing is trying to get CEOs who are on track to gigaton scale, um, technologies to actually have us in mind a year before they need us, which I think for a long time, people didn't even think we were still in business or operational. And so the fact that like, People actually think that we're in business or oper…

AI assessment note: “I think that you're going to see an even split, um, Between the advanced technology”

Q buffer of some storage, but you could basically produce on demand for the power plant. The end, Then you would need to store the hydrogen in the first place. Is the, but the idea here is that this is a peaker plant, and so it's not operating at high capacity factor, and you don't want to have to Run the electrolyzers at the same time that the plant is running?

A Yeah, no, that's exactly right, right? I mean, when you think about, um, using these kinds of electrolyzers and running them all the time, um, you really can get below four dollars a kilogram for, uh, for clean hydrogen, right? And so, um, but that does require you to get affordable electricity, but also very high run times, right? And so, so I do think that the cost would go up substantially if, you You were running the electrolyzers only when the peaker plant was running. I think the other piece of this, though, remember, is that if we're going to treat hydrogen and this sort of high penetration, uh, variable renewable energy grid as something that is a fixture of the future, then you do need this kind of storage that is, uh, really applicable to the entire grid, right? Right, and so when you think about variable renewable energy, obviously you have lower loads in spring and fall, so you have great amounts of overproduction during those time periods, and then you have, um, large amounts of load during summer and winter, uh, for the temperature extremes, right? And so, uh, so I do think that whether it's pumped hydro, or whether it's, uh, hydrogen, or whether it's other things, I do think it's important to recognize that, um, having that level of storage allows for, uh, The more advanced business models to achieve itself. So whether it's overbuilding, you know, more wind and s…

AI assessment note: “the cost would go up substantially if, you You were running the electrolyzers only”

Q rate. So the argument for a battery then, the economics for a battery, the simple version of it is basically just charge up when you're going to have excess generation, discharge that battery into the grid or into the home when prices are high and you're doing an arbitrage. Are you saying that's not like a sufficiently, that's not ascribing sufficient value to the battery, just doing the time arbitrage?

A Yeah, I think, I mean, maybe stepping back for a second, the, the point is, like, what are we solving for, right? Like, I'm not solving for the solar industry to continue to be able to, like, find a value proposition for customers, right? What I'm solving for is that we are sitting in a moment right now where we are being told, and the data is suggesting that it's true, that we're going from 0.4% load growth a year to two to 2.5% load growth per year. Right? And when that happens, the question becomes, like, what tools do we have to be able to meet this load growth, right? And so one of the tools that people talk about is the fact that we have 40% rooftop solar in Australia, and that we should have something more similar to that here in the United States, right? And so that could be one source of generation. And, you know, if we can sculpt it in a way that makes sense to the grid, then, like, it, it, Offsets all sorts of distributions with some upgrade costs and all sorts of other things, right? But we're in this moment where we're having a conversation about how we compensate homeowners, right? So NEM three point O basically says that if you self consume the power, then you get paid all this money. And if you export it to the grid, you get paid very little money. So you need to add a battery to, to maximize your self usage, right? But it doesn't actually force the electric uti…

AI assessment note: “it doesn't actually force the electric utilities in California... to use the batteries”

Q but then with the potential to Expand it and then use a bunch of other things. Do you think of it as being from just like a structural perspective? Um, there's a project developer that has developed ACEs, the storage asset, and then finances everything else around it. Or how do you think about the, the ecosystem of players that have to be involved for this to work at scale?

A Yeah, there are many ways of starting this conversation, and all of them are equally confusing, so we can just pick one, and then we can go from there. I'd say that in Mitsubishi's case here, from the conversations that I've had with them, um, they've been pretty clear about the fact that the salt dome is what they were solving for, um, and that, uh, there are 12 or 15 more salt domes like this around, uh, the West, and so, um, So this is a model that might be replicated. I think that separately, when you think about hydrogen, um, I think everyone acknowledges that moving hydrogen is a pain. Right? Like, whether you're blending it in natural gas pipelines, or whether you're liquefying it and sending it through trucks, or whether you're, you know, just doing compressed hydrogen, or whatnot. Like, moving around hydrogen at scale is not something that we're looking to do at the same level of, um, of, uh, frequency as we're currently doing with natural gas. Right? I mean, and even with natural gas, which is a much easier molecule to handle, you think about methane leaks and how, like, How prevalent they are. Um, doing it with hydrogen would be even harder, right? And so I do think that, um, not unlike low, uh, low-cost hydro and how aluminum plants and others try to, like, you know, co-locate to where those exist, I do think when you think about the industrialization of our country…

AI assessment note: “navigate towards Production of hydrogen very close to the consumption”

Q Or at least faster, right? Even if it's not cheaper, right? To your point on interconnection cues, there's a decent case that you can build the aggregate same amount of capacity behind the meter faster, just because of how long it takes to get interconnected at utility scale now, at least in some places.

A Yeah, I think that's exactly right. So then the question becomes, when you have a system where you have a, you know, Tesla Powerwall or Sonnen system or whatever it is, and there's an inverter that's a part of that, right, and that system can be installed without solar, right, Right. But my sense is it could become sort of solar ready, right? Then that system has its own payback rubric, right? Whatever that is. And now the question is, how much is the solar? You know, and so, so now that I've got a battery with an inverter, right, and it's solar ready, how much is the solar, right? Today, that number is, you know, whatever it is, four 75 a watt or whatever it is with the inverter, right? Right. But if the inverter is already paid for over here, and, you know, now you're doing solar, you could imagine that, like, maybe you could get a lot more competition with that upgrade, that solar upgrade, right? And maybe you can get it down to two 50 a watt.

AI assessment note: “Yeah, I think that's exactly right.”

Q second order problem emerges pretty quickly, which is, like, a peak that occurs a lot more often, and you still have to solve that problem. So, like, having a bunch of interruptible load is helpful. It's better than nothing, but it doesn't solve your problem if you're adding Gigawatts of new load to a grid that hasn't seen gigawatts of new load in a year since the nineties, you know?

A Well, but the reason I, I mean, and you're right, this is probably a different conversation, but the reason why it matters for the virtual power plant piece to me is that these are not loads that are going to get added to the grid in the way in which people think they're going to get added, right? So when you think about how our electricity system has morphed since the Right? Everyone decided to get air conditioning in the 19 eighties. That's why we have natural gas beaker plants and all this other stuff. Right? So as a result, the system capacity utilization of our system has been going down every year since the 19 seventies. Right? And so when you think about all these loads that you're saying is getting added to the grid, um, that does not mean that we have to add The system peak, as you're, I think, insinuating and suggesting, so I think you're right. And, and so the question then becomes, how best do we deal with that? And my point to you is, we have 400,000 megawatts of batteries in interconnection queues. The reason they're stuck in interconnection queues is because the ISOs think that they have to have dedicated transmission such that they can actually charge in the middle of a peak. Right? That's crazy talk. Right? And so they should all be approved next week with certain, you know, like, sort of assumptions around when they can operate and when they can't operate. Rig…

AI assessment note: “these are not loads that are going to get added to the grid in the way”

Q so I should also be part of this VPP. So do you have a view on, just to make this successful large scale, um, Does there need to be a single aggregation point at the building level or the distribution level, or can we just figure out a way for all of these disparate loads and sources of generation and storage and so on to automatically play with each other?

A Yeah, it's so, it's a good question, and one that I think, um, depends on where you end up with, you know, your sort of tribal proclivities. I think, um, there, it is most certainly the case that if you want to backfeed power into the grid, then you will need a much more robust and expensive structure. Right? So if you were to say, like for those, for instance, there's a lot of people who are obsessed with vehicle to grid. Right? If you want vehicle to grid, then you need a very expensive set of permissions from the utility and others before they'll allow you to backfeed onto the grid. Right? And so if you make a, a simplifying assumption, which is that no one is backfeeding anything into the grid.

AI assessment note: “if you want to backfeed power into the grid, then you will need”

Q Because specifically, you don't, the, the mandate, or rather the risk to the loan programs office would be much higher if you're offering a loan to build a manufacturing plant for a company that then can't finance its operations because they can't raise equity. Is that, that why the equity, ability to raise equity is so important from your perspective?

A Yeah, I think there's, there's a lot of people who want the loan programs office to be, you know, sort of, um, your fairy godmother in Cinderella, right? Where you, you're, you, you show up in rags, and, you know, I create a, uh, carriage for you out of a pumpkin, and, you know, turn all the mice into, like, horses, and all that stuff, but it's just not the way it works, right? I mean, in general, we're a liquidity instrument. We're not, A subsidy instrument, right? So maybe our interest rates are more competitive than commercial debt might be, but that's not intended to be a subsidy. It's intended to be, um, you know, market rate debt, and so the question really becomes, you know, if we're a liquidity instrument, how do we lean in when commercial banks are not leaning in, right? Commercial banks don't want to lean into the monolith story, not because, um, they're against the story. I think they see the broader Uh, macro trends around hydrogen. They see the macro trends around carbon black and, you know, Goodyear and Michelin's desires to, you know, be a little bit more sustainable, uh, than they are today. Or they see the macro story around, um, you know, figuring out how to, uh, get equity returns in the marketplace, right? Um, doing these kinds of things. But fundamentally, they're saying, I can kind of meet my numbers by just doing stuff that's easier. Um, Like, do I really…

AI assessment note: “we're a liquidity instrument. We're not, A subsidy instrument”