Why Turbine Shortages Could Slow AI, Data Centres, and Decarbonisation

AI may be booming, but the real bottleneck to it's growth may be turbines. And if firm power can’t scale fast enough, parts of the energy transition hit a wall.

In this episode, I’m joined by Brad Hartwig, Co-founder and CEO of Arbor Energy, to unpack a part of the climate tech and energy transition story that gets far too little attention: the physical machinery needed to deliver reliable, round-the-clock power. Arbor is developing modular supercritical CO2 turbines with integrated carbon capture, aimed at tackling one of the hardest problems in decarbonisation: how to provide firm, scalable electricity while still driving emissions reduction and keeping net zero in view.

We dig into why turbine shortages are becoming a serious constraint on hyperscale data centres, utilities, and industrial electrification, and you’ll hear why Brad believes this is now a critical choke point for both AI infrastructure and climate progress. You might be surprised to learn how stretched the traditional turbine supply chain has become, and why legacy manufacturers may be structurally mismatched to meet the moment.

We also get into oxy-combustion, methane leakage, biomass, carbon sequestration, long-duration storage, and the awkward reality that wind, solar, batteries, and grid expansion, while essential, may still leave gaps when it comes to firm power. This is a grounded conversation about climate tech, policy, energy transition strategy, and what serious infrastructure thinking looks like when the easy slogans run out.

🎙️ Listen now to hear how Brad Hartwig and Arbor Energy are rethinking firm power for a faster, tougher, more honest climate transition.

Sign up to Climate Confident+ for deep dive analysis of the major climate and energy stories of the day.

Transcript

Copy

Brad Hartwig: 0:03

More than 50% of data centres are delayed because of power generation. The, demand for, turbines that can provide power 24 7, is on the order of 120 gigawatts a year. Which is more than double the global production capacity for turbines. That has resulted in a six year backlog for turbines. Turbines are the bottleneck right now for the entire AI industry

Tom Raftery: 0:33

Good morning, good afternoon, or good evening, wherever you are in the world. Welcome to episode 266 of the Climate Confident Podcast. My name is Tom Raftery. Before we begin, a quick update. Last Friday I published the first bonus episode of Climate Confident+, where I took a close look at the Iran War and what it means for the global energy system. That episode is available exclusively to Climate Confident+ subscribers, and it's the first of what will be roughly two bonus episodes a month from here on out focused on major developments in climate, energy, and the transition as they unfold. Subscriptions start at five euro a month, and that support helps keep me producing Climate Confident as a space for serious conversations about climate solutions, emissions reduction, clean energy, and the practical realities of decarbonising at speed. Now onto today's episode, there's a comforting myth in the energy transition that if demand shows up, supply will somehow follow. Build more renewables, add more storage, expand the grid. Problem solved, right? But the reality is rarely that tidy. Sometimes the bottleneck is much more prosaic and much more dangerous. Sometimes it's the hardware. Because right now one of the things constraining the build out of AI infrastructure, firm power, and even parts of the broader electricity system is not policy and not demand, it's turbines. More specifically the ability to manufacture enough of them, fast enough. My guest today is Brad Hartwig, co-founder and CEO of Arbor Energy. Arbor is developing modular, super critical CO2 turbines aimed at delivering firm power with integrated carbon capture. And in this episode, Brad lays out a striking argument that turbine shortages are becoming a critical choke point, not just for power systems, but for the AI industry itself. We discuss why legacy turbine makers may be poorly aligned with the needs of this moment, why modularity and additive manufacturing could matter far more than most people think, and how Arbor's approach fits into a wider debate around gas, carbon capture, long duration storage, nuclear, and what clean base load might actually mean in the years ahead. If you want to understand one of the hidden industrial constraints shaping the energy transition and why the next power crunch may be as much about manufacturing as generation, this is a conversation worth your time. Brad, welcome to the podcast. Would you like to introduce yourself?

Brad Hartwig: 3:30

Thank you, Tom. Yeah, I would love to. I'm Brad Hartwig, the co-founder and CEO of Arbor Energy based in Los Angeles, California.

Tom Raftery: 3:37

For people who don't know, Brad, what is Arbor Energy? What is it you're doing? What problems are you solving?

Brad Hartwig: 3:45

Arbor is developing fuel flexible turbines that use super critical CO2 as the working fluid. And our product is really designed to meet the growing demand of hyperscalers data centres, utilities, for base load 24 7, 365 power on a much faster time timescale than what's becoming standard today.

Tom Raftery: 4:04

For people who might be unaware, Brad, what's super critical CO2?

Brad Hartwig: 4:10

Supercritical CO2 is carbon dioxide at very high pressures and at high pressures, even at room temperature, it behaves in between a gas and a liquid. Because of that high pressure, it is much smaller for the same amount of power. It can be high efficiency. You don't sacrifice performance while making a smaller machine that is fewer parts and overall cheaper to build. And we see this as, a critical advantage because right now when we look at the global turbine supply chain, if you were to get in line for a turbine today with GE, Siemens, Mitsubishi you're gonna be waiting six years or more. We're looking at 2032 as earliest delivery time for a new turbine. And the benefit that we have is our turbines can be additively manufactured. The cause of the bottleneck for those big turbines actually comes down to a few bespoke components. Industrial heavy castings, forgings, blades and veins and the turbine blades are one of the most difficult parts. They're single crystal calfed blades. It's the only way you can get, a strong enough component that won't, fail and creep high temp corrosion. and there's only a couple shops in the world that can make those blades. So we have a, a turbine that can be completely 3D printed and is high performance, but much lower cost while sidestepping that supply chain.

Tom Raftery: 5:40

Okay. Before we get too deep into that, tell me a little bit about the origin story of Arbor. What made you wake up one morning and say, I know, I think I'll start Arbor?

Brad Hartwig: 5:51

Yeah, my background was originally in aerospace engineering and I started my career developing rocket engines at SpaceX, leading manufacturing for engines that now power the crew dragon vehicle, transporting astronauts to and from the International Space Station. I was on the front lines doing evacuations of communities and from wildfires in Northern California through search and rescue work. And that kind of put the energy and climate problem front and centre in my mind. And being an engineer, I was thinking kind of the biggest lever I have is to work on solutions that can help us mitigate the problem. And we really see decarbonising the global economy as one of the, biggest problems of our time.

Tom Raftery: 6:33

And let's talk about the, broader system you're stepping into. How serious is the power crunch you're seeing from the likes of artificial intelligence and electrification? Those are the two main levers we're seeing now that are increasing demand for electricity.

Brad Hartwig: 6:54

Yeah, it's, it's huge. we started Arbor before even this big crunch with, AI 'cause we were already seeing that electrification for transportation, industry, residential is gonna just drive a lot more demand for electrons and increasingly 24 7, 365 power. Electrons are really becoming the, primary currency of a global energy economy. Now enter AI and just the rapid pace of, demand for deploying data centres that consume power all hours of the year. It makes up a relatively small portion of energy mix today, but it is by far the fastest growing piece of energy demand. This year alone, more than 50% of data centres are delayed because of power generation. And we're seeing that the, demand for, turbines that can provide power 24 7, is on the order of 120 gigawatts a year. Which is more than double the global production capacity for turbines. And so year over year, what we're seeing is that has resulted in a six year backlog for turbines, which has, there's an a number of components that are very, inelastic, supply. It's a very large industrial capability that you need to build out and highly specialised labour that you need to build out, both of which add years and years to the timeline to even expand production capacity for traditional turbines. Way we see it is turbines are the bottleneck right now for the entire AI industry, and specifically within the turbines, it's turbine blades. So it's, yeah, it's a huge problem.

Tom Raftery: 8:44

Okay. And just for people who are listening who might not be aware, when you're building a data centre, small data centre, you can probably plug into the local grid, but the larger data centres and the hyperscalers, they generally have onsite generation, and that onsite generation uses typically gas turbines. Which is where we're seeing the problem access to these gas turbines from, as you said, the, the manufacturers is severely limited at the moment.

Brad Hartwig: 9:13

Correct.

Tom Raftery: 9:14

And is it just for that kind of use case, or would you see customers in other industries and fields as well? Or is it just primarily data centres?

Brad Hartwig: 9:24

We have a lot of excitement from utilities, from industry. it's interesting because the state of the gas turbine market was go bigger, bigger, bigger because you get increasing efficiency through economies of scale, going to 300 to 500 megawatt size machines. And outside of the data centre world, 500 megawatts at a single site is a lot of power. So what we see is a 25 megawatt turbine is actually in many ways more flexible than where the traditional turbine industry has been going. We can get really high efficiency. for a 25 megawatt machine, because we're using super critical CO2 technologies. And so we see a lot of interesting applications for local utilities, for behind the metre for industrial applications. Even the ability to do load following where you have maybe a bulwark of renewables plus storage, but you have the ability for a full, seasonal dispatchability with a zero emission turbine. That is what we've seen is the way you get the lowest cost power to rate payers at the utility level is you have these relatively cheap assets that are able to dispatch power to the grid even when you have long periods without sun. Windless winter spells where the wind isn't blowing either. Thermal power plants are quite unique in that way, where they have essentially infinite energy storage.

Tom Raftery: 10:56

And your turbines sit in and around the 25 to 100 megawatt space. Are they then stackable for want of a better word, you know? Can you have, instead of buying 1 25, can you buy 10 25 to get 250 or is that a separate project?

Brad Hartwig: 11:15

Yeah, we, focus on a 25 megawatt Lego. You can think about it and you can just put together as many Legos as you want. If you're building a data centre, if you need 250 megawatts of IT load and a PUE of 1.4, you need 350 megawatts. If you need some redundancy on that because you want 99.9% reliability. It might put down a couple extra units. And that's actually really helpful because it gives you redundancy. It allows you to have that high system reliability that you wouldn't get if you just plopped down a single 400 megawatt turbine. We see it as extremely beneficial for having flexibility in application, being base load, being load following, being dispatchable, as well as that ability to have a, a smaller Lego that's still high performance and zero emission, allows you to kind of build your own power plant based on how much you need.

Tom Raftery: 12:16

Okay. And I guess two questions coming outta that. Why have legacy manufacturers largely ignored that smaller size and the bottleneck, the supply chain bottleneck we're seeing right now, is it temporary or is the turbine industry just structurally misaligned with modern demand?

Brad Hartwig: 12:40

Yeah, it's really interesting, there's a lot of heritage in the traditional gas turbine world where the demand in prior generations, even for the last two decades in the United States, demand has been relatively flat and it's just about driving costs down more rather than rapid expansion. And so what that really kind of led to was how do you just make a turbine increasingly efficient and combined cycle power plants are really good at, using least amount of fuel to get the same amount of power. So it was a machine that was really kind of optimised for the problem at hand. What's interesting about those OEM primes, kind of their business model, is that they have a lot of incentive to keep building the same machine because it's, not really in the turbine sale that they make money. It's in the long term service agreements where they are servicing those turbines. Those are the 20 plus year contracts where they may sell the turbine at cost, but they might get 50, 80, 90% margin on servicing of the machine, because you don't want just anyone servicing your, gas turbine then is providing power to a, city., they are very incentivised to continue with the business model as is. They're also making in incredible margins right now with the supply crunch that is, happening and we're seeing, there's a whole bunch of factors that have led to this, but a combined cycle power plant that used to cost 1200 to 1500 a kilowatt, recent numbers in the new$3,500 per kilowatt to build that plant. So more than double the cost of a, of the plant, the same plants from 10 years ago. We don't see them wanting to get into this space on their own. They're too entrenched that it's too big of a money maker for them to keep going with business as usual. As far as the supply chain bottleneck improving. It kind of depends on your, thoughts on, will humans be ever satisfied with the amount of energy available to us? And generally we've seen that people only want more energy as time goes on. you know, AI compute becomes more efficient and then we find new applications for AI. And so, we just continue to grow, our hunger for energy. It's just really interesting because the demand side, in many ways, the way we see it with AI will always be able to grow faster than how the production will be able to for those large turbines, those large frame machines and those, those primes have actually been burned very hard in the past from over investing in production capacity, like in the, uh, the mid two thousands where the demand actually didn't pan out. And so they actually have a lot of scar tissue from those days where they're hesitant to invest heavily in production capacity or they, they wanna see demand be high year over year, over year before they decide to make that investment to expand. If the pace of demand keeps accelerating, it's always gonna be a lagging, effect compared to where the, the industry is going. So I don't expect demand on the traditional turbine side to really catch up ever, at least not in the next decade. We're keeping our tabs on a whole, lot of factors.

Tom Raftery: 16:10

It's called Jevons' Paradox if I remember correctly, William Stanley Jevons, the Economist from the, the Industrial Revolution, reaching back into my first year college economics classes to drag that one outta my brain there. so. Let's talk a little bit about the technology choices behind your approach. at a high level, you're using oxy combustion. what makes oxy combustion different from conventional gas turbines?

Brad Hartwig: 16:41

Yeah, so it's, it's really this nice partnership between using super critical CO2 as working fluid, and using oxy combustion for zero emission power. And what we see is in a traditional power plant, air is both your working fluid and provides the oxygen for combustion. A traditional turbine operates at around 40 bar, 40 atmospheres. Ours operates at over 200 atmospheres, much higher pressure. If you have a, traditional gas turbine, it's going to, burn fuel with air, and then it will send all of that hot air and those emissions out a smokestack. What we're doing is, we're basically removing most of the components of air on the front end. We're just taking the oxygen and that's done through a standard process. it's using off the shelf area separation units. Process is basically cryogenic distillation. You're cooling the air to pull the oxygen out, and then when you burn fuel with pure oxygen, if you look at, all fuels that, we leverage today, whether it's natural gas, whether it's lower, BTU sin gas fuels their hydrocarbons. And so when you burn, say methane, CH4 with pure oxygen at stoichiometric conditions, you're just gonna get CO2 and water. So your only really combustion products. When you expand those through a turbine, you can then just condense the, the water out and you now have a pure stream of, high pressure CO2 that's ready for direct sequestration. So it gives you, integrated carbon capture, zero onsite emissions. and it's, all on a much, much smaller form factor. We don't have extra compressors to compress CO2 to dehydrate it. It's all part of the standard process, so it allows you to kind of delete a lot of things in the power plant, ultimately driving the system cheaper while also being zero emission.

Tom Raftery: 18:49

And just for, let's say I have a data centre and I place one of these onsite for my generation. What do I then do with that CO2? I mean, I can't put it in the box and put it in the back of my Jeep and drive it off to the dump. What, where does it go? What happens?

Brad Hartwig: 19:10

Yeah, we look at for early on deployments where there is nearby geology for durable storage. And so we look at class six wells are basically wells dedicated to, CO2 storage. So if you think in the case of, right now everyone is grabbing for natural gas turbines because they're able to provide that, reliable 24 7, 365 power. So you can kind of think of pulling the fossil fuel out of the ground, you are combusting it and then all the CO2 goes back into the ground. It's, basically a power plant without a smoke stack. So it's, unique in that way. There are utilisation pathways where you can take that CO2 and turn it into other useful things. usually those pathways end up with the CO2 back in the atmosphere at some point. So this is what we see is kind of dealing with it right at the, source is one of the most efficient ways to ensure that the CO2 never makes it in the atmosphere.

Tom Raftery: 20:08

What about things like then methane leakage? If methane leakage sits above 1%, does saying you've got zero operating emissions still hold up systematically.

Brad Hartwig: 20:20

So we actually are working with partners that offer what is called certified low leak natural gas. Basically you don't want your product leaking because that's just revenue lost. But also from a, a climate perspective, of course, it's depending on how you look at it, 32 to 64 times more potent of a greenhouse gas, having methane leaks. And so that is very central to our procurement strategy on the fuel side is partnering with folks that have certified low leak natural gas, making sure that those upstream emissions are as close to zero as possible as well.

Tom Raftery: 20:58

Talk to me a little bit about the fuel flexibility, because you said methane is just one of the potential, fuels that you can use. What other fuels can you use in this turbine?

Brad Hartwig: 21:38

We actually started with biomass as the fuel input We were seeing actually that there's a huge opportunity to decarbonise natural gas power given this resurgence of demand for AI. but what's interesting about biomass, if you gasify it, you create syngas, a mixture of, methane, carbon monoxide hydrogen, this mixture of gases, but biogenically derived rather than fossil. So you take waste biomass, convert it into a syngas, and now you can burn that in this turbine as well. When you sequester the CO2 from one of those machines, you up there creating carbon negative power, you basically letting the plants while the growing scrub, CO2, from the atmosphere, and then you are creating power out of it. basically solar energy in the plant, and then you're sequestering the, CO2. We're really excited long term about that pathway because it's basically a, a machine with positive externality. It's providing power to the primary economy, but then as a byproduct is naturally putting away CO2. and whether you're burning traditional natural gas or biomass, you're also generating water as a byproduct through that oxy combustion. In the case of, biomass input, it's almost like a zero waste carbon, negative water positive machine. We see it as kinda like a central node to a whole lot of good that could be done.

Tom Raftery: 23:05

And do I need to site somewhere close to where the rock formation will allow me to sequester the CO2?

Brad Hartwig: 23:15

We're promising for all of our early project sequestration because one, if you're doing gas, it's the only way you get low carbon intensity power. and then if you're doing biomass, the benefits of the carbon negativity are huge. So we think there is a lot of benefit to be nearby storage geology. And there basically, I think of the CO2 pipeline network is, in the US maybe where the natural gas pipeline network was 50, 70 years ago. So we expect there to be a lot more investment in CO2 infrastructure such that you don't need to be near the geology, you just need to actually be near a transportation network for the CO2. And that would unlock basically unlimited deployment flexibility. But for the near term, we expect that you wanna be near the geology for CO2 storage.

Tom Raftery: 24:05

And you mentioned you worked on making the rockets for the Crew Dragon for SpaceX. What did aerospace teach you about let's say speed and, and vertical integration, for example?

Brad Hartwig: 24:18

It's interesting. A lot of our team comes from the aerospace world. My co-founder and CTO also came from GE and then, SpaceX after that did clean sheet design of, rocket engine turbo pumps. And a lot of our team has experience in oxy combustion for rocket engine's, turbo machinery for rocket engine turbo pumps. A tonne of overlap in the technologies we're developing here at Arbor. but on the vertical integration side and on manufacturing side, it's everything from how we approached developing the product for manufacturability from the start. We saw how actually difficult given our, CTO used to design the first stage turbine blades at GE that was kind of his bread and butter for over a decade and just how, difficult of a process that actually was. We'd also just spent the last five years, pushing the envelope of design using additive manufacturing. And seeing just how much design complexity you could kind of get for free with additive manufacturing with a process that's far more democratised, and we saw the ability to own those printers, own the design. You really get to control your own, destiny in a way that startups traditionally never could. And so right now we're using other vendors that have 3D printers, just because we're, we're an early stage startup, but the plan is, okay, well, as we get more capital, we will own the 3D printers as well. We'll own the post-processing, we'll assemble the machine. We'll do all of that under one roof. And that was a superpower for us at SpaceX was the combustion engineers were sitting right next to the turbo machinery engineers were sitting right next to the systems engineers and all of the engine folks were also sitting next to the folks designing the structures for the rocket. And they're sitting right next to the folks doing the control systems for the rocket to steer it back to the launchpad. And having the conversation amongst all of those, engineers is incredibly powerful because what we see is, engineering can't be done in a silo. When you're building a complex system, you're constantly making trade-offs, not only within your, piece of the machine, but also with every other piece of the machine. You're working on this, you know, final product of which you're contributing a piece to it, but being able to make trade offs and say, Hey, you know, I could, make my part much, much cheaper or much more higher performing if you over there on the, piping team could, beef up your flange. If you make your mating feature a little bit stronger, then I can use much cheaper piping. And we use a lot of piping in the plant. So all of that can come down in cost. And when we look at the overall plant economics, we see, oh, that's the better investment. Let's, make this part actually a little bit more expensive, because it'll make the whole plant cheaper. It's often at the integration of, subsystems where, programmes die. It's like, oh, well, who was paying attention to that interface? So we saw very early that having as much of the, special sauce, and as much of the, plant architecture in one house was gonna be a major advantage in not just developing a new technology, but making something that is actually going to work and something that's gonna be mass manufacturable. So it's everything from the design, the manufacturing, the testing. We actually built our own test site so that we can test at a much faster cadence and own our destiny for, I think if you go to a traditional national lab, they'll want you to hand over your article. They'll do testing and then they'll give you a, stack of papers at the end and it's like, here's the data from your machine. If we're iterating really quickly, we want our engineers testing the thing and feeding that back to the design engineers, or maybe it's the design engineer is actually the one that's taking the hardware and going and testing it and seeing like, oh, that's, that decision that I made was, was not so good. Like I, I'm feeling the pain of that now in testing. Let me go back and redesign that so that it's, gonna be a better product overall. That extreme ownership and vertical integration where it really shines.

Tom Raftery: 28:39

And what surprised you most moving from rockets to turbines?

Brad Hartwig: 28:45

Yeah. I'd say from a technology standpoint, there's a lot of similarities. it's basically rocket engines for propulsion, but it's almost a rocket engine for power generation. It's funny 'cause that's where turbines came from as well, is it was really a derivative at the beginning of aerospace propulsion for jets. you take that engine and now you apply it to power gen. So we're doing something similar. the hard rule needs to last a lot longer than a typical rocket flight. you know, I think of instead of. a couple dozen minutes for a rocket to, now with reusable rockets where the goal was pushing the boundary of, how long these things have to last. And, that was eventually Elon was pulling people from, GE for a time because he wanted these systems to be able to be rapidly re flown and, and last for a long time. But it's still, quite different than a traditional turbine that operates for maybe 28,000 hours for three years between real overhaul, maintenance for, a turbine. So there's different aspects of materials. What does, good look like? having a system that can operate reliably 24 7 for months on end, years on end is, a unique challenge. The good thing is you can beef things up a bit. On a rocket it needs to be so, so light because it's every ounce or every gramme of weight matters a lot when you're trying to get to space. That's not so much the case for terrestrial machines. You still don't want everything to be a brick because of, cost and, logistics, but it certainly allows you to beef up your factors of safety so that when people think of rockets, I think sometimes they think of explosions, and that is not the case for terrestrial machines. You get much higher factors of safety, make everything ASME certified, OSHA compliant allow people to be operating around the system. things that you generally wouldn't want to have with, with a rocket.

Tom Raftery: 30:57

And I saw an article recently that talked about long duration energy storage, Google contracting Form Energy, for their iron air batteries. A 30 gigawatt, 300 gigawatt hour battery. so a hundred hours of storage in the battery. what do you think long duration energy storage, what role do you see for that in this segment now, given that these kind of times and durations of storage are now becoming economically viable?

Brad Hartwig: 31:33

Yeah. I think it's, in many ways, long duration energy storage is a, we see it as a compliment. When we look at utility scale systems, what we're seeing is each technology in the overall kinda landscape contributes something unique. Solar is by far the cheapest cost of new energy, but it might only on its own operate 20% of the time, and so you can add shorter duration storage lithium ion to get you an extra four hours fairly cheaply. As you start to add more and more energy storage it, it becomes quasi asymptotic in terms of the overall costs, and so that's where long duration energy storage comes in is helping you flatten out some of that asymptotic curve by increasing the amount of energy storage you can have from maybe four hours to what could be a couple of days. And so that's what we see is a, is a huge unlock for the grid. It's difficult because at, at the utility scale you are planning on, Hey, what are the things that are gonna cause outages? Like when you stack all the worst case aspects like on top of each other when you have a, a massive winter freeze and no wind and you're not topping up your batteries, there can be weeks on end when you don't have have renewables available. It's tricky. You have what are basically ancillary services, things that can kick on and provide you power when that's happening. We see thermal power plants are, going to always have a, a role to play in terms of, their ability to firm the grid over all time cycles. Like if you don't generally think of energy storage or duration for, thermal power plants. It's just, yeah, it's always there. It's always available. and then on the other side of it is of course the, pace of, deployment. And when we think about, a thermal power plant like ours, you can get 100 megawatts in a couple of acres. And that is quite unique or different compared to what would be two orders of magnitude, more land required for solar, three orders of magnitude, larger required for wind and not everywhere has that abundant land rights to just deploy, renewables all over the place. It, really depends on how much reliability you need. If you're okay with, 90% reliability across the year, then you could probably get there with long duration energy storage. If you can get it up to scale from production, it'll still be more expensive than what we're doing, but it would be possible, maybe you could get it to 200, $250 a megawatt hour. We're looking at long term getting to 50 to $70, a megawatt hour, levelised of cost of electricity with zero, zero emission. with the ability to operate with three, four nines, so instead of 90%, it's 99.99% reliability. At at five nines, you're talking about only five minutes of downtime all year. And that's what utilities are going after. When they get everyone up in arms about, Hey, where's, where's my power? When you have an outage or what data centres, but a lot of folks want is they generally aren't okay with taking a month off where they just don't have any power.

Tom Raftery: 34:58

Actually I realised after I said it that I misspoke on the Form Energy, battery size. It's a 30 gigawatt hour capacity battery, giving out 300 megawatts of power. So it is still a hundred hours, but it's 300 megawatts of power, 30 gigawatt hours of of storage. If we look forward, let's say 10 years, 2035, 2036, what do you think at that point the likes of clean base load actually means?

Brad Hartwig: 35:33

It's really interesting. I think it in part depends on what future you plan to bet on. right now there's a huge push in the US for new nuclear, small modular reactors for creating clean base load power. It's interesting because the way that you create power from a nuclear plant is you boil water and you create steam and then run that through a steam turbine. So it comes back to a turbine and, a super critical CO2 cycle you don't need to burn anything. You can actually do waste heat recovery. You can, you have a heat exchanger where you are taking heat from a reactor and heating super critical CO2. Actually you can get upwards of 10 points of cycle efficiency higher from CO2, than you could from water. Really that we see it as really interesting, even for, for that future. If you're looking at natural gas and biogas and increasing carbon capture, of course we, have a really exciting role to play in that. Also, there's, interesting forms of long duration energy storage that are thermal batteries where you're basically just heating up a big rock. and then, then the way that you get the heat out is either get it out as, as more heat or you, again convert it into in to power through a turbine We know a number of folks that are working on long duration, thermal storage, where the ability to convert it into heat or power is really interesting with super critical CO2 cycles having a, a unique application there. And so those are being charged by, solar and wind and whatever is cheap and abundant during the day, and then can discharge when power is, needed and those, resources are not generating. I think it will continue to be an all of the above approach. And, energy is really interesting because it always becomes very political and depending on what resources you have available within your country's borders, depending on who you have friendly trade with. All of that factors into the types of energy that you use as a country. So I, I generally don't see a one size fits all approach being how things end up. I think we're gonna see a lot more solar. I think we're gonna see more wind. I think we will see more nuclear. How fast and how big it scales to is a, I think there's large error bars on that. We expect, I mean in the latest IEA reports and, and we've seen that they're now no longer showing a, a hump for natural gas between now and 2050. They just expect the demand to keep going up for the next 25 years. And so we're like, okay, well if that's the case, then we, certainly want to try to decarbonise that. You know, coal is, is coming down, oil, consumption for power gen is, coming down. All of those have seen a peak or are about to. Natural gas just continues to increase through this 21st century. So we expect all forms of energy are gonna be incredibly important for certainly in the next decade, if not the, coming century. So I think having a solution that can play well with, wherever you get your energy from is gonna be crucial. We're seeing in general, the turbine bottleneck is, relevant to a whole host of whatever you choose as your fuel carrier, whether it's gas, biomass, nuclear, all of it relies on a turbine to end up converting that to useful electricity.

Tom Raftery: 38:58

Well, one other form of long duration energy storage that I featured on the podcast a few weeks back is advanced compressed air energy storage. I had on, VP from Hydrostor, a Canadian company who were doing it in the US and they're doing it in Australia and a few other places. I think Germany is another place. They have a plant as well. And again, it's, it's essentially, you take air, stick it underground under pressure when you've got excess energy and then when you need energy, you just release it up through turbines again and generate electricity that way. So very, it sounds very simple, but they have some nice technology on board to help them do that.

Brad Hartwig: 39:37

Where it's applicable, where you have the geology for it, pumped hydro is, one of the more efficient forms of energy storage. Geologic hydrogen storage is something that people are exploring, compressed air storage, if you have caverns, a good underground geology for it. It's not, typically a one size fits all, but it's, I think as a civilization, we're gonna continue being hungry for more and more energy, and there's gonna be a lot of interesting solutions that come, to the surface as a result.

Tom Raftery: 40:08

Quick lightning round of questions. So, one sentence answers. Let, let's see if we can do this. First off. Build fast or build perfectly. Build Fast. Okay. And natural gas with carbon capture, or no combustion at all.

Brad Hartwig: 40:31

I think the reality is we're going to be burning natural gas for a while. We might as well make it zero emission. And that's I think, the most practical approach to what'll be this century.

Tom Raftery: 40:42

Okay. One big power plant are lots of smaller modular ones?

Brad Hartwig: 40:50

Modular power plants give you flexibility in application, and I think are increasingly gonna be how projects get financed and deployed.

Tom Raftery: 41:00

Okay. Are batteries enough to solve firm power in the next 10 years? Yes or no?

Brad Hartwig: 41:06

No.

Tom Raftery: 41:09

If methane leaks can't be kept extremely low, should we still build new gas plants?

Brad Hartwig: 41:17

Yes. I think the demand side people are gonna keep building natural gas plants, so it makes sense to capture all the emissions and to tighten up the infrastructure upstream. It makes more sense to focus on making those plants better, given that people will continue to build them

Tom Raftery: 41:33

And what's the bigger bottleneck today, getting a grid connection or getting turbines built?

Brad Hartwig: 41:39

Ooh it's close, but turbines are currently the, the longer lead time.

Tom Raftery: 41:46

And a left field question for you now, Brad. If you could have any person or character, alive or dead, real or fictional as a champion for your system, who would it be and why?

Brad Hartwig: 42:05

I'd choose probably Carl Sagan. he is, he is been an inspiration of mine. I always think, he's got a line, I'll, I'll butcher it, but it's, it goes along the lines of, we have a responsibility to act more kindly to one another and to cherish the pale blue dot, which is the only home we've ever known, and that is a huge inspiration for what we're doing here at Arbor.

Tom Raftery: 42:28

Okay, we're coming towards the end of the podcast now, Brad, is there any question that I didn't ask that you wish I did or any aspect of this we haven't touched on that you think it's important for people to be aware of?

Brad Hartwig: 42:41

The timeframe by which we plan to get to market or when might you start seeing Arbor systems powering a grid near you.

Tom Raftery: 42:50

Okay. What's that look like?

Brad Hartwig: 42:52

Our plan is to be putting electrons to the grid by 2028 with the system broadly available in the following year. So scaling up both the technology and production capacity very, very closely together. And the goal is that we are, by the end of this decade, we're operating, putting power reliably on the, grid in the US and then expanding from there.

Tom Raftery: 43:17

Great. Brad, if people would like to know more about yourself or any of the things we discussed on the podcast today, where would you have me direct them?

Brad Hartwig: 43:25

Go to LinkedIn, follow Arbor and feel free to reach out. Send me a direct message there. It's Brad Hartwig on LinkedIn, founder CEO of Arbor Energy.

Tom Raftery: 43:34

Perfect. Great. Brad, that's been really interesting. Thanks a million for coming on the podcast today.

Brad Hartwig: 43:39

Great. Thank you so much, Tom.

Tom Raftery: 43:41

Okay, we've come to the end of the show. Thanks everyone for listening. If you'd like to know more about the Climate Confident podcast, feel free to drop me an email to tomraftery at outlook. com or message me on LinkedIn or Twitter. If you like the show, please don't forget to click follow on it in your podcast application of choice to get new episodes as soon as they're published. Also, please don't forget to rate and review the podcast. It really does help new people to find the show. Thanks. Catch you all next time.