The Exchanges

Q along, which is the, the G Hitachi BWRX 300. That's the one that's going to get built in Ontario and maybe TVA territories. That'll be the first one. So the basic question is like, are we going to see a ton of new reactors deployed in the market? Or are we mostly just going to see the one or two that Sort of have already gotten mostly through the gauntlet.

A I think the short story is there already is a nuclear renaissance happening globally, hasn't quite caught on yet here in, in North America or in most of Europe, at least Western Europe. Um, and we can see the answer playing out, which is that there's just a few reactor designs that are getting traction, and basically it's the ones that you mentioned, especially the AP 1000 at this point. Um, Um, and, you know, China is, is very much driving that. And it's actually one area of technology in which China is still, uh, you know, buying a significant, buying a significant amount of technology from Western, a Western vendor. Um, and I think that that same pattern is going to play out in the nuclear Renaissance as much as it happens anywhere in the world. There, there just can't be a Cambrian explosion of new reactors. The, Industrial logic of the nuclear industry just doesn't lend itself to that. I think best case scenario, it's bad for the industry if you end up with, you know, four or five competing reactor designs that are relevant in any given region, because really what you need for nuclear to come down the cost curve is you need, you need, uh, economies of scale throughout the supply chain, and you need to really come down the learning curve, uh, Uh, when it comes to deployment. And I, and I would say that learning curve extends all the way from policymakers and regulators down…

AI assessment note: “There just can't be a Cambrian explosion of new reactors.”

Q And then the other category, the things that are dispatchable capacity, which you should just run through a quick list of, they do give you the on-off button, but they do not generally give you the on-off button as long as you want it bit, right? That is, that is important to note.

A Of course, yeah. I mean, yeah, there, there's mostly no such thing as a perfect energy resource here, and I mean, the closest thing to a perfect resource from the perspective of a, a grid operator is, you know, a flexible, you Um, distributed genset of some sort that can turn on and off really quickly, can ramp, ramp up and ramp down very quickly. And ideally, if it's a natural gas genset and tied into a gas distribution line or a gas transmission line, um, it, it really can, you know, operate just like any other power generation resource and give you an on-off button and, and turn on for, for as long as needed. Particularly If it's a really efficient, low emissions, natural gas genset, uh, for example, you know, one of our portfolio companies at EIP, Enchanted Rock, has been deploying that type of resource for a long time now. Um, and so, you know, it's not restricted by air permitting concerns. You know, that, that's the closest thing to a perfect resource from a grid operator standpoint, because it's, it's just like, uh, a centralized generation asset. It just happens to be spread out there at the edge of the grid. Um, So that's one category is distributed generation, but, you know, some of the, the newer categories that I think are interesting, um, that's, that's one category. There's kind of Three other, really two other primary categories of distributed capacity resource …

AI assessment note: “Of course, yeah. I mean, yeah, there, there's mostly no such thing”

Q can extrapolate to today when the market definitely needs this, and prices, at least for some things, batteries especially, are significantly lower. And so, the thesis of your piece, I think, is maybe the time is right now, and actually this was a real thing, but it was 10 years too early, or whatever it was. Is that, am I framing, like, sort of how you think about it right?

A I think that, yeah, that's exactly right. I'm, I'm pretty darn confident that the time is now. Um, of, of the two big blockers from 10 years ago, I think we've, like, one of them is completely flipped on its head, which is the power grid didn't need DERs that much in 2015. Now it desperately does, and, um, you know, you and I have talked about all of the reasons for that on this podcast before. There's this Electricity gauntlet metaphor that we've been using at EIP for the past few years to describe the state of the market. We, we don't have to be labor here, but like every part of the system has major bottlenecks now. So if I convened that group of, that same group of utility engineers, um, from 2017, if I convened them again today and said like, are you seeing places, hot spots on your network where there's Load growth that's going to be very difficult to meet and extremely expensive to build out, you know, uh, and upgrade a distribution feeder. Like I, I have zero doubt that the answer would be very different today. So that, that's changed a lot. Um, I think it's still more of an open question. Um, Whether and how much we can really move the cost of deploying various classes of DERs down. That's the piece that, like I said, we're, we're starting to see budge in certain, in some ways, but it has been more stubborn over the years. And so, um, You know, the hope at this point i…

AI assessment note: “I think that, yeah, that's exactly right. I'm, I'm pretty darn confident”

Q which expires earlier than it would have in the IRA, but on the other hand, keeps. 45 X, which is the manufacturing tax credits all the way through the early 20 thirties. So you still have But, of course, introduces FIAC, right? So it's like a very complicated equation to determine, is it still worth it to stand up your new, I don't know, battery cell factory in the U.S.?

A It's super interrelated, and that's why it's so hard to parse. But I actually think that battery manufacturing, simulating battery manufacturing, in my opinion, is probably the most important thing that The IRA was trying to do when it, when it comes to making, uh, clean energy manufacturing overall more, more robust and less subject to geopolitical risk intention in the U.S. Um, and One, you know, so I, I think it's worth starting at the beginning, which is, like, can you comply with these FIAC restrictions and make battery cells in the U.S.? And it's tricky. I, I think there's some real uncertainty there, and it really depends on the kind of batteries you're making, because especially if you're making, um, you know, NMC cells, so the, the, uh, more expensive cathode material cells, then, you You know, that cathode active material can be, like, roughly half the cost of a cell. Now, there is supply outside of China. Um, there's some, you know, relatively minimal supply of that material today coming in out of the US, um, with some more on the way. But if you add up that, that cam, the cathode active material, and then anode material, which is much more concentrated in China, that's, you know, graphite anode, um, powders, Um, which is another, say, 10 to 15% of the cost of a cell. Um, then it's hard to comply with fiat if you were, if you were buying those materials from China to…

AI assessment note: “can you comply with these FIAC restrictions and make battery cells in the U.S.?”

Q which expires earlier than it would have in the IRA, but on the other hand, keeps. 45 X, which is the manufacturing tax credits all the way through the early 20 thirties. So you still have But, of course, introduces FIAC, right? So it's like a very complicated equation to determine, is it still worth it to stand up your new, I don't know, battery cell factory in the U.S.?

A It's super interrelated, and that's why it's so hard to parse. But I actually think that battery manufacturing, simulating battery manufacturing, in my opinion, is probably the most important thing that The IRA was trying to do when it, when it comes to making, uh, clean energy manufacturing overall more, more robust and less subject to geopolitical risk intention in the U.S. Um, and One, you know, so I, I think it's worth starting at the beginning, which is, like, can you comply with these FIAC restrictions and make battery cells in the U.S.? And it's tricky. I, I think there's some real uncertainty there, and it really depends on the kind of batteries you're making, because especially if you're making, um, you know, NMC cells, so the, the, uh, more expensive cathode material cells, then, you You know, that cathode active material can be, like, roughly half the cost of a cell. Now, there is supply outside of China. Um, there's some, you know, relatively minimal supply of that material today coming in out of the US, um, with some more on the way. But if you add up that, that cam, the cathode active material, and then anode material, which is much more concentrated in China, that's, you know, graphite anode, um, powders, Um, which is another, say, 10 to 15% of the cost of a cell. Um, then it's hard to comply with fiat if you were, if you were buying those materials from China to…

AI assessment note: “can you comply with these FIAC restrictions and make battery cells in the U.S.? And it's tricky.”

Q from the bill. So we're going to run through five of them that you and I identified before. Let's start with the first one. I'm going to call this one What the FIAC. Um, so FIAC restrictions abound in this bill. Talk me through, I guess, at the high level, the FIAC restrictions. Like, where do they come into play? And then your high level view on what it means.

A I think FIAC is the biggest question mark because it's so omnipresent. You know, it cuts across, like, the, the, the three most important categories, at least in my opinion. Of, of the tax credits that are being modified here, and those are the ITC and PTC for renewable projects and storage projects, I mean, deployed projects, that is. Um, and then the manufacturing production tax credit, the 45 X, all of which are subject to slightly different versions of FIAC restrictions. Um, which, you know, very simply in my mind, I'm just thinking about as, you know, you can't buy stuff that Is originating in or controlled by companies in China, basically. Um, it's more complicated than that. I'm sure, uh, lawyers would jump down my throat and point out all the, the nuance there, but that's the basic gist of it as far as, at least as far as I think I need to understand. Um, and, you know, that has really different impacts across each of those categories, right? So for wind projects, it's not a huge impact because it's very easy to source components from, you Non-China sources. There's plenty of, you know, wind turbine component manufacturing in America and Europe, et cetera. Not a big deal. Um, although, you know, wind has other troubles in the, in the, uh, the big bill. Um, for solar and battery storage, it's, it's more complicated, right? I think, yeah. And, and then for manufacturing, …

AI assessment note: “those are the ITC and PTC for renewable projects and storage projects”

Q All right, let's move on from storage. What's another category where you think the better mousetrap fallacy is rampant?

A So I think the, in my opinion, the now canonical example of the better mousetrap fallacy is in the nuclear fission world, where I think for decades the problem with nuclear power has, has really not been a technology problem. I mean, I've written about this. Lots of people have written extensively about this, and, um, I think it's been a public relations and opinion and regulatory and policy problem, right? We first built nuclear power plants starting in the late fifties and through the sixties and into the seventies, uh, you know, that are, that are operating well today. And we built those plants before we had computer modeling software and all of, all the wonderful engineering tools we have at our disposal today. So nuclear Technology works. We, we used to be able to build it pretty, uh, cost-effectively. Um, and in fact, nuclear power is still being deployed pretty cost-effectively in countries in which it is still being deployed, uh, today in which are, are ramping up deployment, like Korea and China, for example. And so, you know, I see nuclear as a, as a space in which, you know, so-called Gen-III technology or, you Gen three plus technology, which is very much an incremental improvement over the light water reactor designs that have come before, is really, in my opinion, all we need. And at the same time, there are, there are dozens of better mousetrap developers out the…

AI assessment note: “the now canonical example of the better mousetrap fallacy is in the nuclear fission world”

Q All right, so let's make this a little bit more tangible. Like, what is a sector right now where you think we are starting to see or already seeing too many better mousetraps?

A We've been using storage as an example, so I don't know, let's, let's riff on that for a little while longer, because, because energy storage, specifically Stationary grid storage. And, and frankly, I would extend this into next generation batteries for mobility as well. So basically any way of electrochemically or mechanically storing energy, um, I think is an area where the better mousetrap fallacy has taken hold to some degree. And again, I want to be clear, like, I think there are opportunities for that, that 10 X better solution in some areas. Areas within energy storage. But that's definitely one where It's kind of amazing now for over a decade, there's almost a continuous stream of companies I feel like I've seen across a wide range of different approaches to building a, a mousetrap to store energy, um, that have always been in the 20 to 50% theoretically better than lithium ion camp, and have just gotten repeatedly over time crushed over and over. And, and, you know, this is an area where I think, um, you know, I mentioned earlier, I, I, I'm concerned that in addition to wasted capital, there's also a distraction, uh, effect of, of having dozens and dozens of mouse traps continuously entering the market in the industry. That's one where I, I believe I've seen that, especially because in, in my role at Energy Impact Partners, I work with lots of large incumbent, uh, ener…

AI assessment note: “energy storage, specifically Stationary grid storage”

Q what makes it different, because you have You have a market that is clearly subject to the better mousetrap fallacy. Lithium ion just keeps getting cheaper and cheaper and probably will continue to do so. And yet we find a stationary storage chemistry that we find attractive. So was, what is it that distinguishes form or just use it as like a way to describe what falls outside the fallacy?

A Right. I, I remember having this conversation with you initially, probably five plus years ago now, and, and we've had a similar version of it over time as, as we've talked to and, and assessed more, uh, better mousetraps within the storage universe. And the basic thrust of the conversation was we, we don't want to be caught investing in a company that is anywhere in the realm of what lithium ion will be. Will approach from a competitive standpoint. And so what was attractive about form is that the fundamental cost of the materials that go into their battery and the way that they're building the battery enable them to build this multi-day system at, uh, at a target cost that, that is so far beyond what lithium ion has achieved today and beyond what any reasonable vision of the future of the technology would entail. That it is, I'd say, outside of that competitive threshold. It is a, it is a different category of product that this incumbent solution, lithium ion, and even, even, you know, steps from lithium ion that are, that are related like sodium ion just won't be able to touch. And another example that's also in our portfolio is thermal storage. Um, Rondo is a company that we've invested in that is a, you know, a thermal storage technology developer that's, You know, pretty rapidly moving into the commercialization phase, uh, of its life cycle, which is really exciting. Um, …

AI assessment note: “fundamental cost of the materials that go into their battery and the way that they're building”

Q do, but, but I worry that because there are so many different options, like, let's just say the, the NRC starts licensing reactors left and right all of a sudden. Would be a total sea change in that market. But now we have 20 licensed reactors. Is that good or is that bad for the future of nuclear in the United States relative to if the NRC licensed, like, two?

A I, I think it's probably bad, sadly, for the future of nuclear in the U.S. I mean, I would obviously love to see, uh, changes in the NRC, which made it, uh, more efficient and, and much less costly to license a new design, but I, I, I think it's much more important that the NRC process becomes more efficient and less costly for licensing projects as opposed to for licensing technology, right? And, and I do think that, you know, if I, if I were the, uh, omnipotent, uh, energy czar of the country or the world, I would pick One, two, maybe up to three designs, and deploy them serially, uh, and, you know, say we're gonna do 10 of these, and we're gonna see how it works out, maybe 10 of each, um, and we're gonna space them a year or 18 months apart from each other, and I almost guarantee that that would lead to much faster cost down, and maybe even just ultimately lower cost than any of the next generation designs that we're seeing, Could really achieve. I, I should, I should, you know, I should, uh, soften that language. Maybe not any, right? But then the vast majority of the, of the better mousetraps that are, that are out there taking up, uh, attention.

AI assessment note: “I think it's probably bad, sadly, for the future of nuclear in the U.S.”

Q Okay, and so as we talk about those things, I think it's still worthwhile to start with the presumption of electrification, right? And then, and then sort of knock down the places where you can't electrify. Do you, do you agree with that? And, and if so, why is that a presumption that's like a reasonable place to start?

A I, I think that we should presume electrification for transport, and, and should always be asking why not electrify. Um, because, you know, frankly, just because of the cost. Um, Compared with pretty much all of the other options for, you know, decarbonizing moving vehicles, um, at this point, electrification looks relatively cost effective, and, and that's because of a couple different factors. You know, one is just because of the kind of the basic cost of the, the fuel itself, clean electricity, and even clean electricity, given the, the sort of pressure on electricity costs that we're Anticipating that we've talked about on prior podcasts and because of the electricity gauntlet, even as the cost of clean electricity increases over the coming decade or two or three, I think electricity will still look cost effective as a fuel, like the primary, you know, the energy input relative to things like biofuels or clean hydrogen, perhaps some other alternatives we can talk about. Electrofuels, for example. And secondarily, kind of compounding that cost advantage is the efficiency of a battery electric drivetrain. You know, battery electric drivetrain can probably get into the realm of sort of 80% efficiency when you're talking about from electricity input into the vehicle to the wheels or, you know, turbine moving of that vehicle, right? And so, um, and that's, that's just, you know,…

AI assessment note: “I think that we should presume electrification for transport... frankly, just because of the cost.”

Q And that's, that's just slow charging the drayage trucks, right? You could want to fast charge them. You could also add The trucks that go in and out of the port, they're not dredged. Like it, it, you know, to truly electrify the entirety of the port would be a massive undertaking from a grid infrastructure perspective.

A That's right. So, um, you know, again, Port of Long Beach is a good example just because they've, they've thought a lot, a lot about this and put out, put out a lot of public material on it. And, um, back in 2011, uh, that port started out on this, this 10 year effort to electrify, uh, Um, all the cargo handling equipment, and also to provide what's called shore power to ships that are basically parked at the port, um, at one of its major cargo terminals, and they have seven major cargo terminals at this port. And it was basically a, that project alone, just for cargo handling equipment and shore power, that took 10 years, um, and required the utility, Southern California Edison, to build a new, Um, 66 kilovolt, kilovolt trans, uh, transmission line and four new substations into the port area. So, you know, it just gives you a sense of the kind of infrastructure challenge we're up against. So, you know, ports are in, in my mind, because of that, an interesting example of where you can theoretically electrify the vehicles, but it might make more sense to use a different approach.

AI assessment note: “That's right. So, um, you know, again, Port of Long Beach is a good example”

Q Which, by the way, makes them perfect for electrification, right? Like, if you're going to electrify heavy-duty trucks, drayage trucks are the way to go, because they don't Go long distances. They go, maybe they make a lot of short trips.

A That's right. From, so from a density standpoint, uh, you know, battery density should not be a constraint for those types of trucks. And actually the economic value proposition should be really high because there's lots of start and stop operations, which batteries and electric vehicles in general are really good at. Right. But infrastructure becomes a concern because if you're going to charge those 1600 drayage trucks, you know, In and around the port, somewhere within 50 miles of the port, for example, which is where they generally tend to go. And even if you were to charge them overnight at, you know, a hundred kilowatts, a hundred kilowatts each, so kind of slow charging overnight, you still have 1.6 gigawatts of new, new peak demand sort of in and around that port. And, um, in this interesting study, the port itself put out Uh, if you look at the existing substation infrastructure, electrical substation infrastructure in that region, that would be adding 90 megawatts to every substation if you had to sort of, ah, use the existing footprint of those substations for charging, which is just completely infeasible, right? So we're talking about adding, you know, multiple transmission lines worth of new power, feeding the port in one of the densest operational environments You can possibly imagine. And, you know, again, it's, it's possible we could eventually get there, but tha…

AI assessment note: “That's right. From, so from a density standpoint, uh, you know, battery density should not be a constraint”

Q And in this case, it's more about volumetric density than about It's gravimetric, right? We don't, we don't care so much how heavy the batteries are, but they take up too much space.

A That's right. I mean, graphic, gravimetric density affects the economics of fueling because you have to push more weight through the ocean, but the amount of space that the battery takes up is really what's a killer because that affects the economics of how much cargo you can carry and how much you can get paid for. And then, you know, given what we were just talking about in, in the context of electrification of Ground transport at ports. You can, you can only imagine what kind of additional electric infrastructure would be required to power, you know, large, uh, cargo ships at a port, uh, and not just providing shore power while they're parked, but actually giving them enough energy to, to make a journey across the ocean. It's just, it's just not practically feasible. So again, except for pretty small, um, ships and, you know, some ferries in certain instances that, you That don't really account for a large share of global shipping emissions. I, I've pretty much ruled out electrification and shipping as well.

AI assessment note: “the amount of space that the battery takes up is really what's a killer”

Q is the charging capacity one, which I think is like less talked about a little bit, but is obviously directly related to all the things we've talked about before on this podcast, the electricity gauntlet. So like, as, as your thinking has evolved on electrification of medium and heavy duty, or I guess Heavy and very heavy duty things. How much does the charging constraint come into play for you?

A I think it's a really big constraint. I mean, I think there are some areas where it's a much bigger constraint than battery density, pretty much everything on the ground, I actually think, including, you know, the heaviest duty, uh, at least road trucks, perhaps not like big mining vehicles and other big off-road vehicles, but like even class eight semi trucks, I think, Have a plausible pathway to maybe not a hundred percent electrification, but pretty high levels of electrification, um, over the coming decades. But that's an area where in some places, despite the fact that you can, you can achieve sort of technical viability and economic viability at the vehicle level with a battery electric drivetrain, if you have enough vehicles that all have to come together to fuel in a We're talking about just enormous amounts of power. So the, the sort of, uh, canonical example, I think at this point is ports. So, um, the example I, I, you know, I know best because I, I, they've put out actually a good amount of data on it is the port of Long Beach in California, which is basically right next to the port of Los Angeles together. They sort of make up the largest port in the United States. And so if you just take Long Beach, sort of half of that combined port infrastructure, There's, uh, the, the port itself put out a study on this topic, and they have about 1600 trucks. They're called dra…

AI assessment note: “I think it's a really big constraint.”

Q starting to see as utilities are being forced to reckon with this new load growth. What is it that they're planning to build? Because this is one of the This is one of the side effects of the pace of the gauntlet's arrival that I think we're going to have to be talking about a lot over the next few years. Like, what does this mean for new generation types?

A It's, it's a real challenge for decarbonization timelines, um, which is what I've been worried about for a long time now. Um, the first thing that utilities are planning to build to, to manage their entry into the gauntlet is they're planning to To not retire their coal power plants on previously agreed upon or decided upon timelines. So there's been already a handful of reports of utilities that had planned on shutting down a coal power plant by 2030 or even sooner, and basically having to say we're going to keep that plant online longer than anticipated, in some cases with no, uh, deadline for, for retirement in sight. Because they, they just need that power. Like they can't retire that plant safely and reliably given the demand that they're seeing. So Evergy is one example, um, in Kansas, uh, where they have a new Panasonic battery factory coming online. Um, and it's been reported that, you know, because of that Evergy has to keep this coal plant running longer than expected. First Energy is another good example. Um, so that's the first thing is we're going to see Coal staying online longer than we expected. The second thing is natural gas, right? So for any electric utility, for any load serving entity, grid operator who has to meet rising demand in the next, let's call it two to three years, even up to five years, the easiest, most straightforward thing to do at this point…

AI assessment note: “the easiest, most straightforward thing to do at this point is to build new gas fire”

Q If so, then kudos to you, Plug Power. Well done. Um, okay, so microgrids, one, one potential, as you said, not small, but specific sort of outcome of this. Others that you think are worth noting?

A I think the other, the other big category that really emerges from this near-term pinch in, in the gauntlet, this really narrow passage that the utilities have to get through, um, is Is, is new gas generation assets, basically. I mean, if you, if you do need to serve gigawatt scale new loads within, like, let's say the next three, maybe even up to five years, it's really hard to find, to think of another option that will really work, especially if that's, you know, more than very low capacity factor kind of peaking type of load, which, which storage is a pretty good, uh, solution to today. Um, But if you're going to build new natural gas power generation assets today, you, you need a way to future proof them. Or I, you know, I think of it as kind of carbon future proofing them. Um, and that can come in a couple of forms. One is, you know, fuel optionality, the ability to build something today, like a new gas plant that can be relatively easily and cheaply retrofitted to burn hydrogen or ammonia, um, over time. And, and fortunately that's something That I think the natural gas turbine OEMs have been very proactive about, um, in developing products that are starting to meet that kind of roadmap where you can build something that is, you know, low level hydrogen, hydrogen capable today, you know, capable of blending in, say, 30% hydrogen by volume. But if you build the plant today…

AI assessment note: “the other big category that really emerges... is new gas generation assets, basically.”

Q power from? And then there's the T and D side, where are we, how are we going to deliver the power? So let's talk about the challenge on the generation side first. What do you see as being the reason why we can't just flip a switch and turn on All the generation that we would need in order, in order to serve all this load, you know, virtually overnight.

A If only it were so easy. Um, I think what we're seeing is that this new big opportunity for the industry, which is a fantastic opportunity that we've just been talking about, is coming at a tough time for the planners of the power generation and transmission and distribution system. On the power generation side, there's really kind of a pinch that these system planners are feeling, and that pinch is coming from a few different factors. One is Already planned coal retirements, coal power plant retirements. Another factor is, you know, new carbon policy, and even non-mandatory but very public net zero power generation commitments that a lot of utilities and other power suppliers have made. So on the coal retirement side, there's, there's still another 50 gigawatts of coal-fired power generation capacity in the U.S. that is Basically on track, already well planned to be retired by the end of the decade. Now, some of those retirements might get pushed out, but I think for the most part, a lot of that retire, those retirements are fairly locked in. And then meanwhile, you know, over the past decade plus, we've been able to retire coal-fired generation capacity, which is what the power grid was basically built around, because we've been able to substitute it with new gas-fired generation capacity. For the most part, although there's, there's a lot of nuance and complexity there, that…

AI assessment note: “pinch is coming from a few different factors. One is Already planned coal retirements”

Q energy storage that will be required beyond just the fact that we are trying to smooth out the peaks and valleys of intermittent generation, but also for the reason of resiliency basically in this system. And so I guess just to contextualize it relative to that 47 days that we've got a fossil fuel inventory, uh, how much do we currently have in batteries, basically, like non-fossil fuel-based energy storage?

A Uh, it's, it's essentially negligible if you're talking about batteries. I mean, just to contextualize in general our sort of renewable energy supply versus the amount Of energy we store in the form of fossil fuels. All of the wind and solar power generated in twenty-twenty-one, if you were to bottle all of that energy up and, and store it in sort of the same way that we store fossil fuels, it would be only about five days worth of energy supply. So we currently store more energy in the form of natural gas underground and in pipelines than all of the energy we generated from renewables, you know, Two years ago. Um, and then if you look at storage on the grid, the ability to sort of store electricity, powerful sort of power to power storage cycles, the biggest resource we have available to do that today is, you know, what everyone knows as pumped hydro storage, where you basically use pumps to move water back up from below a reservoir to above, uh, a dam into a bigger reservoir, uh, and then let gravity do its work one more time again. Um, we have about, In terms of primary energy supply, not in terms of electricity generation potential. If you were again to compare that to primary energy supply, we have about nine minutes worth of pumped hydro storage, which means we have probably, I don't know, in the seconds level of battery storage capacity today.

AI assessment note: “it's essentially negligible if you're talking about batteries”

Q sure to add, you know, I think there, I think it's important as we start to talk about other technologies, not to forget Why lithium ion is a powerhouse, so to speak, in this context, but, uh, but it does have some limitations, and limitations that pertain to particular applications of energy storage, uh, especially. So yeah, what do you think are the sort of core limitations of lithium ion?

A The limitation is basically that there is, there is a floor to lithium ion and sort of related, uh, related battery chemistries to the dollar per kilowatt hour cost that you can achieve. And I, I don't know exactly what that floor is. No, nobody does. Um, because we are going to assume that there's going to be changes in anode and cathode materials over time. I'm sure changes to To some degree in, you know, the, the full module design that will probably continue to cut costs out of the system, right? Um, but the floor is, in my opinion, probably somewhere north of a 150 dollars per kilowatt hour installed. And I would say almost certainly north of a hundred dollars per kilowatt hour total installed cost.

AI assessment note: “The limitation is basically that there is, there is a floor to lithium ion”

Q Okay, but so you mentioned, um, using hydrogen in gas turbines then. Is that a part of your skepticism, or is that a part of your optimism?

A That's actually become, um, one of the reasons, I think, for optimism for using hydrogen, uh, as an energy storage medium, which is that, you know, the, the gas turbine fleet, even starting today, is, you know, in the early stages of being set up to be Retrofit down the road in order to run on higher and higher blends of hydrogen, and this is partly because utilities and grid operators, they still need new capacity today, and they need new capacity today at gigawatt scale while Frankly, there is no energy storage system today that is ready to confidently deliver the scale of capacity they need to balance out the renewables that are being added to the grid already today, and probably for the next, let's say, five to eight years. And so we're going to be building a bunch of new gas turbines, and as utilities are building these new gas turbines, they are considering the fact that If they're going to run for a 25 to thirty-year useful life, they need a pathway to be gradually, increasingly decarbonized over time, and maybe fully decarbonized at some point. And the, the natural gas turbine OEMs, there's really, you know, sort of a big three turbine OEMs, GE, Siemens, and Mitsubishi, are basically offering them the answer to this conundrum, which is, you can buy these retrofit-ready Gas turbines today, which, which can currently blend somewhere in the ballpark of 10 to 18% hydrogen b…

AI assessment note: “That's actually become, um, one of the reasons, I think, for optimism”

Q talk about energy storage in all of its various glorious forms. But before we get into sort of the nouveau type of energy storage, let's talk about the type of energy storage that we already have on the grid, which comes in the form of basically stockpiles of fuel for fossil generators. So talk a little bit about how much of that we have today and like, what's its importance?

A Right. I mean, You know, when people talk about renewables, oftentimes the, the kind of big question they ask is, how are we going to, how are we going to store enough energy to balance out the variability of wind and solar? And the answer currently is we already have plenty of energy storage capacity on the grid today. It's these massive stockpiles of fossil fuel that we have sitting around. We've got big piles of coal sitting beside coal-fired power plants. We have, well, this isn't as, as relevant to the power sector, not really relevant to the power sector, but In terms of storage supply for the energy sector writ large, we have these giant tanks of oil, um, ranging in scale from big industrial facilities, and of course, we all store oil in our vehicles today, in our, in our gas tanks. Um, and then also very relevant to the grid, we've got these big continent-spanning networks of natural gas pipelines and giant underground storage reservoirs, and so we have, we have all this energy sitting around, ready to be utilized at a moment's notice, and that's particularly true In the case of natural gas, uh, when it comes to power generation, because it can be utilized relatively quickly. If you add it all up, it's actually a pretty large amount of storage. We, we currently store, uh, by my rough count, about 47 days on average worth of primary energy supply in the country. So that'…

AI assessment note: “We currently store, by my rough count, about 47 days on average worth”

Q be clear, total installed cost is an important metric because you see lots of numbers quoted about either cell cost or module cost for batteries. This is inclusive of that, plus all the balance of systems, plus labor and installation, EPC margin, right? Like, what is the turnkey cost of a stationary storage system? So that's what you're saying is probably ultimately north of a hundred dollars a kilowatt hour.

A Yeah, I believe so. Um, that's, that's where I would, I would feel comfortable, uh, making a bet. On, on a competing technology. Somewhere in that hundred to 150 dollars per kilowatt hour range. Although obviously I'd, I'd love for it to be lower. And the reason I'd love for it to be lower, um, is not just because of, uh, the fact that that makes it more secure against future competition from whatever comes out of the mammoth lithium ion industry. It's also because that's, that's what you need in order to make affordable grid storage for much longer durations, right? So Again, because as you add more hours of duration, you add more kilowatt hours per kilowatt that you have in the battery, um, you need, you know, if you want to go up to a 12 or 16 hour, uh, storage solution, which can address really, you know, the diurnal differences in supply and demand, uh, in the, in the power system, particularly that we're anticipating as more and more renewables come up, come online, then you need just a, a significantly lower Battery cost.

AI assessment note: “Yeah, I believe so... Somewhere in that hundred to 150 dollars per kilowatt hour range.”

Q messy process of getting an accurate quote. You can't do it remotely yet. You know, people, the consumer awareness is low. So it feels to me like there's, there's a need sort of across the value chain from, you Building a better heat pump to figuring out how to get it in customers hands faster, easier, and with a lower acquisition cost. Do you, do you agree with that thesis?

A I agree. As you know, I've been, I've been trying to understand how much of this is a, a product problem, and technology problem, and how much of it is a kind of go-to-market sales, financing, packaging for the consumer problem, and I guess what I've come to is those problems are actually more interrelated than I initially understood, um, and that's because these Uh, getting to a, a more efficient, higher performance heat pump would reduce some of the sales friction and would reduce some of the objections that installers and, and customers ultimately have to them. I mean, to begin with, if you could increase heat pump efficiency, uh, particularly, you know, in relatively cold temperatures, um, without dramatically adding to the cost. So if you could find a way of, you know, Of improving performance and driving the, the capex down for that additional performance, then heat pumps are just going to look on a total cost of ownership basis, more attractive, more competitive relative to continuing to burn natural gas or oil. Um, and then also if you can make heat pumps more efficient, uh, at colder temperatures and you avoid the, the winter night problem That you mentioned earlier, where there are periods in the winter, and especially in the middle of the night where it gets very, very cold, that, uh, heat pump efficiency plummets, then maybe for a higher percentage of households, Yo…

AI assessment note: “I agree. As you know, I've been, I've been trying to understand”

Q Okay. And so that, that's the cooling side, but I think that's where there's like a little bit less kind of novelty from a heat pump perspective, right? Because again, this is not a new thing. On the other hand, in heating world, uh, heat pumps are less ubiquitous by a long shot today than, than they are in cooling. So talk to me about heating.

A Yeah. So, um, obviously today, the way we heat buildings, the way we heat water, in the vast majority of cases, um, in northern climates where it gets cold, um, Is we burn fossil fuel. We burn natural gas predominantly in North America today. Um, in many parts of the country, in the world, though, we still burn, you know, fuel oil. Um, and part of the reason is because, uh, the equipment is relatively cheap. A furnace or a boiler is a really low upfront costs, relatively speaking, in terms of installing and, and getting a heating system. Um, and You know, also because historically, and, and continuing on through today, it's just, it's more difficult to make heat pumps work in a heating setting, especially in relatively cold climate environments, and the one reason for this, um, is because of the differential in temperature that you're trying to move heat across, uh, and that's a major driver of the efficiency of a heat pump. If you're trying to move, uh, heat from A, uh, seventy-degree environment, uh, outside to a sixty-seven-degree environment inside, then that heat pump is going to be an extremely, uh, efficient, you know, cooling system, essentially. Um, and if you think about even kind of worst-case scenarios for air conditioning today, say an air conditioner in Arizona needs to move heat, um, From about 70 degrees, let's say, uh, which is a relatively comfortable point fo…

AI assessment note: “it's more difficult to make heat pumps work in a heating setting”

Q going to be, have we caused a new set of problems for the grid? Because we love electrification, obviously, but not all electrification is purely beneficial electrification. It causes a bunch of side effects you have to think about. So what would it look like to see heat pump adoption for, for heating and cool climates at scale here? How big a deal would that be from an electricity perspective?

A Let's start with an individual house, and again, think about in a, in a cold climate like up here in Maine, you know, what it looks like, um, with, you know, heat pump technology that, uh, if you have a really, really cold snap in the middle of the winter, you know, again, this winter night problem, and the temperature gets down to -15 degrees for, I don't know, 8:10 hours, um, potentially even longer, what does that do to electricity demand for a single house? And in short, You know, roughly speaking for a typical U.S. household, it means adding new peak load to that home that's equivalent to about, ah, two electric vehicles that are plugged into level two chargers, ah, during those cold periods. And that's on top of what we hope to be an electric vehicle or two, or two that are, that are actually plugged into level two chargers at the home. And so what this might do in cold climate environments is, is Basically shift peak demand from the summer to the winter and increase the peak by about three X. And so, you know, it, it is certainly possible that, uh, electrification is sort of a pure winning strategy that we could, we could find a way to stage upgrades of the grid all the way from the distribution system to transmission and build out generation to serve that new peak load, uh, But I think it is a big challenge, even if you have, you know, pretty significantly more efficien…

AI assessment note: “shift peak demand from the summer to the winter and increase the peak by about three X”

Q we globally, or at least in North America, take deep decarbonization pretty seriously over the next decade. Um, and the things that can be electrified start to get electrified. The things where there is some alternative pathway, those pathways emerge relatively quickly. What then is the role of natural gas? Like where, Where are we still using it, and how is it shipped around in, I don't know, 1520 years?

A Yeah, I mean, I think, um, there's a, there's a scenario in which, in a very rapidly decarbonizing world, probably the, the best preservation of use for, for natural gas and gas infrastructure, which are, you know, abundant, low cost resources that already touch, you know, most end users in the country. Um, I think one of them is through Probably, like, I'm, I'm actually very excited about this turquoise hydrogen pathway or a blue hydrogen pathway that, um, decarbonizes gas, natural gas, um, at the end of, you know, or at the point of consumption from large, large-scale pipelines for industrial facilities. Um, and I think we could certainly see that being a significant pathway for, Industrial decarbonization alongside also a significant amount of electrification and, and potentially a significant amount of green electricity, I'm sorry, green hydrogen pumped, piped in from elsewhere. And then where I'm pretty confident we'll, we'll see the longest tail of natural gas for the foreseeable future is in all of the end uses that are currently served by natural gas at the distribution level. So again, That's, that's predominantly building heating. Interestingly enough, well, while it seems innocuous, we don't think about it all that much in the scheme of, of kind of big energy end uses. I'm kind of convinced that building heat ends up turning out to be one of the toughest to just sort…

AI assessment note: “longest tail of natural gas for the foreseeable future is in all of the end uses”

Q Okay, distributed energy resources. I don't want to spend a ton of time just like laying out all of the different types of distributed energy resources to start, but I thought you had kind of a useful categorization in this article that you put together. So start by categorizing, like how do you separate out the different types of DERs?

A So, uh, in this article, I have a whole taxonomy, but we want to get into the details. Um, I think there's really, there's really two key categories of DERs, and, and they're the things, basically the things that are distributed capacity resources, which I can define in a second, and then the things that are not. And maybe it's easier to start with the things that are not, which is energy efficiency and solar are sort of the two primary Uh, categories of DER that don't really provide you dispatchable capacity if you're a grid operator. And I don't want to, I don't want to throw shade on energy efficiency and distributed solar, because I think those can both be extremely valuable resources, including for grid planners, not as much for grid operators, because they don't give you real, they don't give you enough real-time control, which is the real challenge here. But for grid planners, they can be extremely valuable. We should be Considering them as a part of integrated resource plans for the power system, we should very much be counting on them, pushing the boundaries of them, etc. But the, the challenge is that what grid operators need increasingly today are resources where they give you a button, an on-off button, so that you can provide capacity to the grid When it's needed, and ideally for however long it's needed. And the problem with energy efficiency and solar is that whi…

AI assessment note: “there's really two key categories of DERs... distributed capacity resources... and the things that are not”

Q Okay, so I'm going to wrap up by asking you to describe two scenarios for me. Five years from now, first scenario, it goes right. And this DERs finally take off and start to reach the promised land. Uh, and they, they solve a meaningful portion of the, of the gauntlet problem. What does that look like? Like, how does, how do we get there?

A To your point, I think we need to start soon because these things take time, uh, and they don't take time because of DER deployment, by the way. That can happen very quickly. That's one of the advantages of many classes of distributed energy resources, you know, whether it's a natural gas genset, um, or whether it's a residential battery, you know, you can, you can go out there and put a bunch of them out there very rapidly, and that can scale up to hundreds of megawatts very quickly, particularly compared to, The bottlenecks in large-scale resources we're seeing today, but it's, it's the programmatic side of things. Like, whatever entity is running the procurement or the program, let's, let's start with the utilities, right, who sit in the middle of all of this, um, they need to start ramping up their programs today if we want to have, um, you know, meaningful, the kind of meaningful, uh, resources deployed within five years that you're talking about. So, I think what we would need to see is lots of utilities saying, uh, you know, look, we, we need capacity today. It really, truly is an all of the above strategy today, and we are going to set up a crack team to go figure out how we get distributed energy resource capacity most efficiently, and we're going to set a goal, and it's going to be, you know, a reasonable goal, but we're going to include that in our plan, in our integ…

AI assessment note: “we're going to include that in our plan, in our integrated resource plan”

Q And then the other category, the things that are dispatchable capacity, which you should just run through a quick list of, they do give you the on-off button, but they do not generally give you the on-off button as long as you want it bit, right? That is, that is important to note.

A Of course, yeah. I mean, yeah, there, there's mostly no such thing as a perfect energy resource here, and I mean, the closest thing to a perfect resource from the perspective of a, a grid operator is, you know, a flexible, you Um, distributed genset of some sort that can turn on and off really quickly, can ramp, ramp up and ramp down very quickly. And ideally, if it's a natural gas genset and tied into a gas distribution line or a gas transmission line, um, it, it really can, you know, operate just like any other power generation resource and give you an on-off button and, and turn on for, for as long as needed. Particularly If it's a really efficient, low emissions, natural gas genset, uh, for example, you know, one of our portfolio companies at EIP, Enchanted Rock, has been deploying that type of resource for a long time now. Um, and so, you know, it's not restricted by air permitting concerns. You know, that, that's the closest thing to a perfect resource from a grid operator standpoint, because it's, it's just like, uh, a centralized generation asset. It just happens to be spread out there at the edge of the grid. Um, So that's one category is distributed generation, but, you know, some of the, the newer categories that I think are interesting, um, that's, that's one category. There's kind of Three other, really two other primary categories of distributed capacity resource …

AI assessment note: “There's kind of Three other, really two other primary categories of distributed capacity resource”