Ceres Power Holdings plc (CWR) Earnings Call Transcript & Summary

Good afternoon, ladies and gentlemen. Welcome to the Ceres Power Holdings plc Investor Update. [Operator Instructions] I'd now like to hand over to Director of Corporate Communications, Elizabeth Skerritt. Good afternoon.

Many thanks. Hello, everyone, and thanks for joining us. My name is Elizabeth Skerritt, and I head up Investor Relations for Ceres. Many of you will also have met our Technology Delivery Director, Jon Harman or, at the very least, seen one of his many publications and podcasts on the topic of hydrogen. Earlier this year, Jon's team announced a collaboration with AtkinsRéalis to design motorized production systems for green hydrogen. And today, we are delighted to welcome Gareth Richardson, who heads up Low-Carbon Technologies at Atkins to discuss the projects in a bit more detail. And then joining them is Chris Leonard, who's the energy transition research analyst at UBS. He's kindly agreed to host the conversation and try and get to the bottom of where green hydrogen is likely to play a role in the energy transition, which industries will benefit and how Ceres' technologies can integrate into these solutions. [Operator Instructions] For the purpose of today's session, we are sticking to the topic in hand, but those that are not picked up will be addressed later by the team. So without further ado, I'd like to hand over to Chris. Thanks, Chris?

Thanks, Elizabeth. So yes, I'm Chris Leonard from UBS research on our energy team covering energy transition enablers across Europe. And it's a real pleasure to be invited to moderate the discussion today between Gareth and Jon. They're both experts in their fields, which means I have the easy job of asking a few questions and steering the discussion, which I'm sure will be educational for all. To kick off the discussion, it would be great to turn to Gareth to introduce himself and the work that AtkinsRéalis does in green hydrogen.

So welcome, everyone. I'm Gareth Richardson, low carbon technology lead. I'm the engineering manager on the work we're doing with Ceres. I've got good in-depth background on what we're looking at. And in terms of AtkinsRéalis as a company, it's a global engineering company and project management with 40,000 staff worldwide. We work across pretty much every sector from infrastructures such as national grid upgrades, major road networks down to building semiconductors and manufacturing plants and mining sites. We even have our own nuclear reactor design, so quite a broad spectrum there as a company. But in the global hydrogen industry, we're supporting -- we're actually running one of the U.S. Department of Energy hubs, the Pacific Northwest Hub, in the U.S. That's about to receive $1 billion of funding from the U.S. government, and we're expecting that to unlock about $6 billion worth of private investment in the area in Pacific Northwest. Some of the projects that we work in, in some of the green space, we're working on green and blue ammonia production, e-methanes. We're in Quebec. We're just part of the design for the e-methane part there. Green steel, we've done a number of projects there. And in the U.K., we're looking at gigawatt hour salt cavern storage up in the north of England, and we're doing the designs for that, including the designs for the caverns. These projects are around the U.K., North America, the Middle East, and we do a [indiscernible] in those from being the [indiscernible] engineer on the client side to actually doing the design all the way through to actually commissioning an ammonia plant -- building and commissioning an ammonia plant in [ the ] last year. In terms of the technology space, we get -- we do -- we are actually a CTO for a venture capital firm. It's dedicated to aerospace. We're doing a bit in the hydrogen area there. We work on new technologies in blue hydrogen and obviously the project we're doing with Ceres to make green hydrogen. So we're really focused on the energy transition and how we can really support that from a technology delivery aspect. There you go, Chris.

Perfect introduction across the whole range of the hydrogen delivery system, so that's really helpful. And over maybe to Jon to introduce Ceres' view on the hydrogen industry today and where Ceres is right now.

Yes. Thanks very much, Chris. Hello, everybody. So our technology -- so really, we started about 5 years ago, 2019, thinking about how we could apply our technology not only for fuel cell applications but also for producing green hydrogen. And we see a really great fit with the technology in terms of integrating it within industrial processes. So if you can do that, if you can use fairly low-grade heat from industrial processes to raise steam, then actually, you can get some really high efficiency, really low electrical energy consumption for generating hydrogen. So we certainly see our solid oxide technology as a really good fit with integrating industrial processes such as steel production, ammonia, synthetic fuels, synthetic aviation fuel and e-fuels as well. And that's really what we've been working with Gareth and the team at Atkins to look at how does how does our technology scale up for industrial applications and how can we get the best out of it to decarbonize those hard-to-abate sectors.

So you guys both announced in February this year that you're working together on a modulized electrolyzed system design and maybe a really good time to just give us a view on how the project is going and what the scope of the work was and is.

Yes, sure. Maybe I'll take that to start with and then I hand over to Gareth. So we -- initially, we started with a technology demonstrator system, which you may well have seen a fair amount on LinkedIn and in the press. So we decided in 2021 to build a 1-megawatt class electrolyzer technology demonstrator. We got that up and running last year. And now that is at Shell's Technology Center in Bangalore just being commissioned. And that's a really great first start to really understand how our technology is going to work at a system level. But thinking about how the technology is going to scale up and scale out is a much bigger question really. And for that, we thought Ceres was best partnering with an organization like Atkins to really think about plant level architecture really from the outside in, so thinking about integration with electrical systems, downstream hydrogen offtake, and then how do you integrate all the different plant level services together and what does it mean to scale up the technology into building blocks that can be made at factory level but then deployed 10-megawatt, 100-megawatt, 200-megawatt blocks. So that's really what we started to look at with Gareth and the team earlier this year. I'll hand over to Gareth for a bit more info on that.

Yes. I think one of the key steps really is bringing those 2 ends. You understand what the requirements are of the base system, the heart of it, the electrolyzer, the stack and then bringing it. Well, what does that end user really need? I'm trying to bring those 2 bits together with all that knowledge so that you actually get a product that the industry wants rather than forcing your impression of it. You understand what the end user wants and then really drive the design from that. And that's what we've done. That's part of this project.

And as a follow-up on that, I mean, obviously, it's fairly new, a few months into the project. But what have been the key learnings and outputs from this early stage of the study? And how does that align with what the end users are looking for. Maybe, Gareth, if you could continue on that.

Yes. I guess, if we started with what the end user wants and we go about what some of the learnings we've come out of it, I mean, one of the key bits is electrical efficiency. You need to have high electrical efficiency and that's where you see it allows you to do that. And as Jon alluded to earlier, when you got steam available, you are swapping out some of that high-value electricity for low-value steam. It still has a cost, but it's a much lower cost than it is using electricity. So it allows you to really drive down that -- the electrical cost of the system. And then when you look at there's a capital cost [ at the end ], they want low OpEx, low capital cost, and that's how you get to low-cost hydrogen. So what we've really focused on is how do we design the various bits of the plants so that you can scale it vertically. And when I talk about vertical scaling, it's not different to how you make the stack. So stack, you're making in multiples. You're making lots of them like cars, right? You're mass manufacturing. And you get that and -- the learning curve on that, and you can get your costs down on that. But if you look at big industrial plants, they haven't gone down that route. They go, right, well, I need to make everything bigger. I need to vertically scale. And a good example of this, if I have a pipe and the pipe, I make the diameter twice, we'll actually get 4x the flow, not twice the flow just by adding the diameter slightly bigger. So it's not like doing sort of horizontal scaling as I call it. You're actually using where you say I have 4 -- I had -- need 4 pipes instead of just 1 that's twice the diameter. So you go find those bits in the plant where you can really scale it bigger and get the cost saving from doing that, from vertically scaling it rather than doing the mass production piece. And that's what industrial plants do. They go, well, I'm going to make my reactor half a meter bigger and get 4x the flow or 3x the flow through it and get production up. And I got more flow and products that averages out CapEx across. So this is what you're trying to do with this, trying to get that CapEx down, high production rates so that you can average out CapEx across more production. And that really helps drive down the capital cost by doing that with the plant. So the actual stack of SAM, so stack array module, SAM, it's core technology. It's a stack that intrinsically, like I said, is mass manufacturable. And kind of the -- there's limitations on how big you can make that. But once you take that away, what else can I do to the rest of that plant? And you can drive the efficiencies there, I think. And I think that's one of the learnings you've got, is there is a lot of opportunity to do that. And I guess coming back to one of the other points about doing that is that where you have electrolyzers that are designed, say, 10 megawatts with all the advanced plants and everything, just multiplying those up on a site, developers are not a big fan of that when they go look at that in the market. The feedback we've had from certain developers is that I don't want to go manage 100 10-megawatt electrolyzers and 100 pumps [ at ] plants. I want to go manage as little equipment as I can because, otherwise, I've got 100 pumps, 1,000 valves rather than 100 valves, 100 pumps that I might need. So it's that -- it's finding out all these different bits from the end user and bringing that into the design to get something that really meets the requirements at industrial scale. You've got to remember, hydrogen is used in industry now. It's not really used anywhere else, so the scale is big. A steel plant needs hundreds of megawatts of capacity. It's 120,000 tons roughly of hydrogen for, let's say, 2 million ton steel plant. So that scale really brings to you -- well, actually, that's not a big steel plant either. So that gives you the scale that you're going to get, just multiplying up lots of little of electrolyzers. Well, the advanced plants isn't going to work for that because your maintenance cost, your operating costs are going to go through the roof. So you really have to find where you're going to get that benefit at that scale and bring that back down to your core design and say how big that module is going to be. And we've done that. I think we've got to the right level module. And then we're scaling up the balance of plant life to meet that at the larger scales. So hopefully, that gives you a bit an idea about what we've learned from the project so far.

Definitely does and maybe a good moment to have a look at what the modules look like. And I think we've got a short video to show the output of the work, and maybe Jon will guide us through that if we can show that.

Yes, Sure. Just a bit of a -- this is a bit of a sneak peek of the video, and it's providing an example of a 100-megawatt plant. So we just -- we thought it would be really great to be able to show everyone this today, but as I say, it's the first cut of the video. I hope you enjoy it. [Presentation]

That's very informative. And I think some of the key takeaways from that might well also be the -- obviously the efficiency gain but equally, the footprint that is very impressive. And I just wondered, when you're looking at the key technology asks from your perspective, customers across the different industries of ammonia, maybe steel or synthetic fuels, what sort of are the key asks and the key challenges of the current solid oxide electrolyzer technology? Is it that scale out to prove it can go to 100 megawatts and 1 gigawatt upwards? Is it the cost per megawatt? Is it the footprint size, which I've just seen there? Or is it maybe reliability? Like what are the key buckets that you guys need to tick to gain those, the trust and the orders?

Well, I think it's definitely they need to be a reliable supplier of low-cost energy. That's absolutely key. And it's got to be like I mentioned earlier, efficient CapEx, OpEx and footprint, as you mentioned, is very key as well. A lot of people forget that land costs quite a bit. It's quite a -- you can look at some feasibility studies, but land costs quite a bit. And if you want a brownfield site as well, so an old steel plant or something like that, where they're trying to build a new site, they can be quite site constrained as well. So having a smaller footprint really, really helps that. I mean a lot of industries are looking for 97% availability of the kit they're trying to bring in for hydrogen. But it will be very high liabilities in that area. And having a modular system that we have allows us to try and get to that 97%, and that's what we're targeting. The other thing to think about in these areas, when you're talking about the industries that we're looking at, big ones, they didn't like going up and down in terms of their throughput. So they really like being steady, which means you need to always have that hydrogen there. So the plant needs to be able to do that. You're really going to have to put large amounts of storage in, like salt caverns, or you're going to have to put a very stable grid connected to it. And again, it also comes down to economics as well with those kind of things. You're trying to achieve at least 65% uptime on everything because you're not going to have enough product to average your CapEx across. So these projects, they're big and they require a hydrogen plant that can deliver the low-cost hydrogen in the right footprint with the right efficiency, the highest efficiency you can get and trying to really drive down that CapEx to meet that. In some of the areas where you can drive down that CapEx, and it's alluded to in the video, people forget that the balance of plant is actually quite a large chunk of that cost and a large footprint as well, which adds to it. And you reduce that with having a higher efficiency by -- the substation now is much smaller, as I mentioned in the video, but also by cooling much longer or losing as much heat and one of the real benefits that -- before we don't talk about enough or people talk about solid oxide is the fact that it doesn't -- and over its life, most electrolyzers degrade, and they have to put more power in across its life to keep the same hydrogen production rate. Okay. Now when you -- with a solid oxide, you have an extra handle, an extra variable you can use to adjust that by increasing the temperature in the system, which reduces resistances. It reduces your electric consumption. So you can maintain the same efficiency by adjusting the temperature of the electrolyzer. So if I'm designing a plant for alkaline or PEM, I need to oversize my substation by 10%, 20%. I need to oversize my cooling by 10% to 20%. I don't need to do that solid oxide. And that is a major benefit in terms of the footprint, the capital costs and the efficiency of these systems. So that's why it's really -- that's what we see is a really great tie into industrial facilities where you have that thermal energy that you can replace your electrical [ actually ].

Yes. Thanks, Gareth. I think you covered that brilliantly. I think I'd just add what are people looking for as well from solid oxide. Well, they're looking for the availability to buy the electrolyzers as well, right? So this is something where, I think, Ceres' particular business model is a real strength. We're licensing business model. So we license the use of technology for people to build into products, but we also license the manufacturing. And we've been working with our licensees developing factories for many, many years, which is scaling up the production of the sales and stacks to many, many, tens of megawatts currently with plans to go much larger. So actually, the manufacturing readiness of Ceres' technology is there as well, so it's ready to hit the ground, really, in terms of scaling out the technology, building into products and then moving into projects, green hydrogen projects in a way that we've really brought the experience of our licensees to bear in this regard to really get the technology to a point where it's great quality, high yield and optimized in terms of its manufacturing cost, too. I think there's probably some other points to just embellish on what Gareth said. So one of the key questions is the delivery pressure of the hydrogen as well comes up quite often. And typically solid oxide technologies whether it be working on fuel cell more, they're operating around atmospheric. They don't need to operate at pressure particularly. And if you just take the technology and run it in reverse and don't pressurize it, it works absolutely fine. But actually, you get a very low pressure out the electrolyzer. And if you're going into an industrial process, that causes some problems, right? So it causes problems in terms of that first stage of compression because you're going from a very low pressure up to the next stage. That's expensive in terms of energy requirements to compress. It's also expensive in terms of CapEx. Those low-stage compressors are really quite expensive, and they're big, and they take a lot of maintenance actually. So one of the key voices we heard when we were developing our architecture was Gareth, was try to deliver the hydrogen at pressure. It doesn't need to be particularly high initially, but even something around 2 bar gets you over that first hurdle to entry, really, those first barriers to entry in terms of getting electrolyzer systems into industrial use.

Yes, I think the bulk of the energy needs for compression is in the first 2 bars. So you get a huge amount of that. And then the curve after that flattens out. As you go up in pressure, there's not as much of an electrical efficiency benefit to doing it. So yes, it's definitely a big benefit there.

And whilst we're still on the compression angle, how difficult is that to achieve in the electrolyzer design, getting that 2 bar to become available? Is that a challenge which you guys feel like you're still early on the development curve for? Will it be challenging for others in the industry? Obviously, it's still a pretty small nascent electrolyzer or sold oxide electrolyzer industry? Will others struggle to get there? Could this be a differentiator?

I don't think it's necessarily a differentiator. I mean, we've -- you've seen in the video those sort of copper-colored pressure vessels. So we took a decision to actually not ask the stack to withstand the pressure. We're installing the stacks within a pressure vessel. And there's other solid oxide designs, which are following a similar approach out there. But I think other companies who may well be set down a certain part in their system designs already and are effectively just reversing those system designs, that's going to be quite a challenge because they're already heavily invested in those factories to build the systems the way they are. So I think it just depends really, but I think we believe that actually having pressurized SOEC for industrial applications has got to be the way to go.

And Gareth, you also spoke about the uptime, to 65% uptime and above, and obviously, you would need to see utilization being high. Is there anything on this -- you spoke about storage, mainly salt cavern storage. And you didn't mention batteries, but clearly, grid connection might be important to enable solid uptime. Has there been any other element of the design that's needed to be sort of modified to enable solid oxide to stay at high levels of utilization? Or is it responsiveness is actually at a enough level to be able to move to swings in renewables?

Yes. So one of the -- we filed the requirements [ piece captured at the start ] from Ceres' partners and industry was big. It doesn't need to be able to vary with the power source that's coming in. You'd have that ability to do that. So the system designed to be able to vary with renewable load, as it goes up and down, and it can do that in a similar way to the way alkaline electrolyzers do it. So it enables you to respond quickly to those. And I guess the point I was trying to make earlier about those kind of things is, yes, you want that, but you don't want to be doing that all day long because if you're varying your power down low all day long or varying it hugely, well, your production rate's going down. Power equals hydrogen, right? So most big industrial sites, they want to be economic, have to run at quite high utilizations to actually do it, which means your electrolyzer has to run at high utilization. Now you might want to be able to vary it at times to do things because there are certain things you can do. If you have a good storage, you can place the services to the grid in terms of demand response, and you want to be able to do that. So having the ability to have that in the electrolyzer design or counted for really helps with that in terms of being applicable across the market. I think probably just add one point to what Jon mentioned earlier about the advantage of business model that Ceres has. There's another advantage to that if you think longer term. Having multiple companies -- other companies building stacks really helps you with the ongoing maintenance. There's access across the market then. If one company goes down, well, another company will have the production facility there building them. So if you have like a one-stop shop, a small electrolyzer company, they don't have that back up there in the longer term. So that's another advantage that we haven't really ever talked about actually that Ceres' business model really helps with distributing that long-term risk in terms of how to change out because don't forget, electrolyzers stacks, like I said, is great, so you need to replace them in the life of the plant.

Obviously, speaking about how essential sort of waste heat will be to improving the power consumption of the solid oxide electrolyzer. And maybe, Jon, it would be helpful to quickly run through and recap why it is that the solid oxide has such a strong hand to play for industrial decarbonization and why the technology is unique with the high temperature aspect to it.

Yes, sure. I mean really, its party trick is operating at high temperature, the stack itself, the core stack. I mean our operating temperature is between 500 to 600 C. We are lower temperature, solid oxide. There are other generations of the technology operate at sort of 750 up to up to 900. But really, the key efficiency is driven by being able to electrolyze steam. That's really your thermodynamic advantage there. That's where you get the most bang for the buck. So if you can feed steam into the electrolyzer, you're already at the stack level 30 -- sorry, 34 kilowatt hours per kilogram at system level, 37, that kind of range. So compared to alkaline and PEM who are low 50s at best, it's quite a significant gap there. But one of the misconceptions, you need really high-grade heat to operate an SOEC system, and that's just not true. As long as you can have steam 150, 170 degrees C from an industrial process, then that's fine. And actually, what happens in terms of getting the steam up to the temperature required to enter the stack because you have a system of heat exchanges in the electrolyzer module we saw in the video earlier. There's a centralized heat exchanger pack and that -- they do the job of increasing that steam temperature to the end of the stack. But that's the point. And really, those sort of grades of heat, 150, 170, 200 C at industrial plant level, they're generally on the way to a cooling tower. There's not much you can do with that grade. If -- normally, most of the useful work has been taken out of those streams by the time they get to those temperatures. So normally, you'd be sending it to a cooling stage or something like that. So actually, that's quite a good advantage of solid oxide. You don't need -- you can use those waste streams and you can generate real value from them.

I'd probably add, there's an important piece on that around the types of industrial technology that we talked about earlier, about fertilizer, about steel and about e-fuels. If you take fertilizer and exothermic reaction, so it's producing heat as the nitrogen and hydrogen are reacting together. And just on its own, the ammonia plants will produce 75%, 80% of the steam -- of the energy required to generate the steam for the solid oxide and you make good fertilizers. So you're running a nitric acid system as well, that's exothermic and you get all the energy you need to produce it. And if you're doing a greenfield design, well, then you're actually removing capital cost from that ammonia plant because now you haven't got a big power generator. They normally have a big turbo generator to recover that heat. So you generate steam with a big steam generator anyway. And then they put it through a turbo generator to recover a couple of megawatts of electricity there. Well, actually -- and it needs to do that. What's more efficient to take that energy is to get straight into a solid oxide to make more hydrogen rather than taking that electricity and putting it to an alkaline plant. So it's another really key aspect. You can't just think about the electrolyzer costs. You need to think about, well, actually what are my full greenfield plant costs or my integration of my full plant costs for that industrial use case, so additional savings you can get there with solid oxide. You're on mute, Chris, I think.

I am on mute. There was always going to be one of those. But Jon, you mentioned the lower temperature aspects of the solid oxide technology from Ceres. And maybe I just wondered if there was a run on for -- when you come to the plant design, what type of metals or materials need to be used if you're running maybe a higher temperature version of solid oxide operating those 750 to 900 Celsius range you spoke to. Does that change the cost? Does that change the balance of plant you need, which obviously Gareth spoke to how important balance of plant is in the total cost?

Yes, absolutely. Yes, we do see a bit of a differentiator there with our operating temperature range. They tend to fall in sort of industrially recognized temperature ranges for materials for example. But also the thermal expansion and also the heat loss is a really important consideration. So with higher temperature solid oxide technologies, you're going to be into quite exhausting materials, quite high-cost materials. And then you want to closely couple the heat exchanges to the stacks. So that tends to lead to an architecture, where it's quite very distributed. You've got lots and lots of heat exchanges all closely coupled with quite expensive materials. Ceres' operating consumption means that, actually, you can have one big heat exchanger pack, which is serving lots of those stack array modules. You can have reasonably long pipe runs between those things. They could be insulated easily. The thermal expansion isn't too bad. And if you do all that, lower cost materials, much bigger heat exchanges, fairly standard design practices, then, of course, the CapEx costs reduce quite significantly in that regard.

Yes, I think one of the key bits was the strength in the steels that we're able to use. So we're able to go down to the 300 series type stainless steels because once you go towards the 650 temperature range and going into where some of the higher temperature SOECs run. Your strength of the material really drops off a cliff. It's not linear. I goes wumph, falls away. So you trying to keep it in that range allows us to use those cheaper steels for a good partner design. It's really helpful.

And whilst we're on technology comparison, maybe pretty useful to speak about the views potentially from Gareth here about the different technologies at play, PEM, alkaline, solid oxide and the landscape of these different technologies. Where do you think solid oxide compares at the moment? Obviously, it's still nascent and it's being developed. But how do you view the landscape at the moment, Gareth, on the different projects you're helping with?

Yes, we always put SOEC into our analysis or most of our clients ask us to do that as well. There's definitely growing interest across the developments in solid oxide. They see the benefits of it. And whenever we do it, we're immediately always coming up with -- in fact, every single time, we're coming up with 10% or greater reductions in the LCoH, typically even more than that. And that really, really means that the developer interest is really huge in this area for the -- especially people that are developing ammonia plants, which drive a lot, when the majority of hydrogen goes [ through that ] oil refineries. So they're really looking for that. And one of the key pieces that they ask or one of the reasons why they tend to go, oh, actually, I want to go with alkaline is technology maturity. And if you look at alkaline, a couple of hundred megawatt scale facility we built in the past, so people know this technology. Now the issue with that is those that liked alkaline electrolyzers in the past will run flat out of hydro. They weren't varying up and down. And one of the new ones that's been built in China, I think it was for [ Sinopec ] or something like that, that was -- I think it was over 100 megawatts as well. They had some issues with that. They had -- when they were trying to vary it up and down, they had gas crossover. And every new technology you would bring into the market, you come across these things, especially when you change the use case. And I think the key thing about Ceres' technology is that they've really got a lot of it demonstrated already with their fuel cell side. So it's actually a lot further on than some developers think. They think, oh, this is completely new coming out of the lab. It's in not. It's in the market now. But as to fuel cell, what we're doing here is changing it over to work in the electrolyzer form. So it's convincing people and having that credible path to demonstration, which is what Ceres is developing and doing. Getting this into the market is really key so the developers feel comfortable and especially people financing these projects, right? It's the finance people that are driving a lot of these technology decisions. They're going I want -- I would say I want nonrecourse finance. I don't want to take that risk. I would much rather you go with our alkaline system, even though it might mean that the project is 10%, 20% better off from a cost -- from operating cost perspective. It doesn't mean that I want to take that risk. So it's trying to get the market ready to take that and take some of those risks. And the market is getting there now I think. There's some bigger and bigger solid oxide deployments going out there and as -- from other companies, and Ceres has got their 1 megawatt and ready to get to the next bigger one, the multi-megawatt scale. It will really demonstrate to the market that once you've got that core building block demonstrated, you can just repeat it then. And then you just got a little bit of risk, but that risk is on plant that you well know. It's the same kind of plant you have in oil and gas facility, so the risk is much lower when you go to the balance of plant. So it's just getting to that demonstration of that core piece.

And Jon, is there anything you wanted to add on the solid oxide market at the moment and capacities that you're seeing?

Yes. Well, I think I'll just come back to our business model again because I think it's -- it provides for further levels in the supply chain than even fuel cells do, right? So in the fuel cell market, we would license to maybe a 7 stack manufacturer, but they tend to also be the system in the product company as well. And you look at examples of our current partners. So they'll be making the components but also design [ limited ] systems and then selling the systems on the market. Whereas in SOEC, you're going to have 7 stack factories, and obviously, that's high-volume ceramics manufacturer making many, many cells a year. Then you've got the kind of industrial process equipment manufacturers that you've got -- you saw on the video, the stack array modules. They're quite large, right? So they're going to be building factories, but they're sort of akin to very large engine sort of scale and the electrolyzer module itself. And then, of course, you've got companies who are then going to be the overall project engineering company such as Atkins and other EPC companies who are then going to be working out the best way of deploying all that technology and the rest of the plant to provide the best solution to the client, whatever their application is, whether it's the hydrogen for steel or ammonia or synthetic fuels or whatever. But -- so I think the -- our business model provides a really rich business opportunity across those different areas. And then that's what we're seeing at the moment, is really strong interest right across that value chain of companies who are interested in moving into that space with a proven technology, proven technology in fuel cell. We've been working on it over 20 years in fuel cell mode. We're effectively operating it in reverse. But a lot of the atmospheres, the temperature, the material selection that it's working in are the same pretty much. So really, we're in a phase of now maturing and optimizing but not -- it's not new invention. It's not starting from scratch. It's a very mature technology basis, which we're just moving into a new application area.

There we go. Sorry. The unmute wasn't working for me. And Gareth, I mean, when we're speaking about derisking these projects and getting first investment, large investment decisions into solid oxide, I wonder if you can maybe elaborate on the potential balance of plant savings and the big areas or the big building blocks of where you would see the save come for the project developers, so maybe on the renewables that might be needed, the lower levels of installed wind and solar, potentially talking about the substation, which would be the point you made earlier. Just really round that point home would be quite helpful.

Yes, I think so. I think you make the point -- yes, it leads to -- you made the points on the saving in electricity consumption and the saving in electricity across the life of the system as well, so hammer that bit home. And if you think about it, the -- one of the biggest costs, you're doing completely greenfield. And one of the things is a lot of these projects require the renewables to be built. Those renewable costs are the same as the big industrial plant. They cost a lot to build the renewables. They're saving 30% on the renewables. Actually, overall project cost goes down massively by having that efficiency versus alkaline or the current alkaline and PEM. And one of the other key bits is the construction cost. The construction costs are -- if you just look at the module costs, the electrolyzer module costs, typically, it's 1.8, 2x that to get you to what a facility cost is. And that facility cost, a big chunk of that is actually construction phase. So being able to minimize that by making the plant modular really helps. Now the way LNG facilities -- and this is a kind of analogous piece where LNG has got to a stage where it's cookie-cutter plant design, where they have these huge modules, 30 meters by nearly 100 meters in terms of area and weighing tens of thousands of tons and they're just picking them up, building them and dropping them on to site. Now what we've looked at with Ceres is, right, okay, what is our mass repeatable unit that I need to get and I need to make that very transportable. And then what you can do is you can build up these huge modules, if you wanted to, to drop them on site. So it's changing that change in thought process about the whole design of the electrolyzer really helps drive down cost and make it more applicable to the developer. So we -- that's some of the bits that we came to.

Maybe we could turn to open up to some of the Q&A that has come through on the line -- well, through the platform. And first, moving to the recent contract that you announced or Ceres announced with Shell to design that 10-megawatt pressurized module. And can you just give some more color on the work that's been going -- will go on for Ceres here? What exactly will Shell be paying for? And what else might you be contributing to that contract? Yes, any color there?

Yes, of course. So as you're aware, we're testing our first-of-a-kind technology demonstrator at Shell in Bangalore. And as a next step, effectively, we're leveraging the work that we've done with Gareth and the team, taking that starting point as a reference design, but then transferring that to Shell to get their input on that. Now they're looking at particular use cases and end uses of the equipment. They're also looking to bring their own differentiation to their projects as well. And that's really how we work at Ceres. We'll take a reference design, a starting point that we've done a fair bit of work on a fair bit of thinking around, but work with our partners to see how they can bring differentiation to the design. Maybe they'll have some suggestions about how best to design the stack array module or the electrolyzer module or how many of those are required or what the overall plant design's going to look like. So it's a first step in that direction. It's a 6 months design study to get to a point where they may well then take a further decision to move forward with Ceres into something a bit bigger and a bit more tangible.

Okay. Super clear. And another question on the module side. So building in units of 8.6 megawatts in size that will scale up to give the 100-megawatt offering. I know, Gareth, you did mention during your remarks that maybe there is a limitation on size for how big the stacks can go in solid oxide. I just wondered if -- and there's a question that's come in. Will you build in 10-megawatt modules? Or will you keep at these 8.6? Will that 8.6 gravitate up towards 10? How are you guys thinking about the system and the modular design?

[indiscernible] There is definitely -- you're able to go bigger than 8.6, right? There is an ability to increase the size of that module. I mean, at the moment, we've got -- there's some 12 SAMs in that module, and there's ability to add more to that. There will come a limitation at a point where the pipe book gets so big, that you then have to start using things like expansion bellows and things like that. And we're trying -- we've managed to avoid that so far. And that's kind of where we got to with the size of that. Now if you increase the size of the stack inside by, say, going from 30 to maybe 40, 50 gigawatts within then the same [ vessel ], you're then suddenly going well into the 10 megawatts, 12 megawatt size with those 12 SAMs. So there's a bit of a constraint in terms of the pipe expansion, those other things, but there are ways that you can increase that module size even further. I think Jon might want to talk about that. But the piece around the stack really helps them.

Yes, absolutely. So there's flexibility there. I mean we perfectly kept the electrolyzer module rectangular. If you wanted to add some more stack array modules or for whatever reason, then you could. There's also a question around redundancy and maintenance and taking modules down for maintenance on overall plant level. So we were looking -- for that particular study that we showed the video around, we're looking at 100 megawatts, say, 12 electrolyzer modules, thinking about if you had to take 1 down for maintenance, what would you be okay with in terms of a reduced capacity for a period of time. And that was a bit of a consideration in that overall concept. But there's no real limitations. I think -- again, from a -- with a stack, what we're trying to do is keep the stack platform common at the moment. So licensees with factories who are making SOFC stacks could use the same factory to make SOEC and the cell footprint. There's lots to be said for commonality in that approach. And that's really why we ended up at the stack array module because SOEC stacks tend to be, just by their nature, smaller than alkaline stacks or alkaline stacks are enormous, right? But we were packaging stacks into the pressure vessel there. And then for the right design study, that's 720 kilowatts in a pressure vessel, and basically, that's sort of like a super stack, right? So it looks exactly like a stack to a process designer, and it doesn't really matter what's going on inside it, to be honest. It does exactly the same thing. It's just got higher power input, higher hydrogen output.

Yes. I think one thing to note about or people think about it is the 720 is quite a little bit close to a megawatt when you compare it to alkaline in terms of production capacity. So you'd have to put a megawatt-sized stack in there. So the 8.6 megawatt of electrolyzed capacity that we've got there is equivalent to a much bigger alkaline plant in terms of production rate, easily over 10 megawatts. So just to keep that in your minds when you're thinking about it, don't -- you always think about the megawatts. The megawatts are meaningless in terms of production. The production kilograms an hour is what's key because that's what you're really trying to put into your facilities. Just thinking about that when investors are looking at it is not one for one when you're comparing alkaline and PEM, you've got to look at efficiency.

And alongside that, is there any comment you can make maybe on the sort of the current density, I mean, equally playing into the footprint being now much smaller? I mean compared to alkaline or your most basic sort of atmospheric alkaline, I would consider that the solid oxide has a higher current entity in terms of maybe amps per centimeter squared or amps per meter squared against the alkaline guys. But I don't know if you -- I mean, you have previously given that analysis. I just wonder if that is the case.

Well, so you can get into a bit of an academic conversation around cell current densities and it's all quite interesting. In fact, I mean, the Ceres -- because we operate at lower temperature than high-temperature SOEC, we have a lower cell current density. But then our cells might be a bit closer together in the stack. And to be honest, by the time you've designed them into the array modules and the electrolyzer modules, that cell current density gets massively diluted into the sort of footprint of deployable equipment that we were looking at earlier. And it doesn't really -- by the time you've got to that scale, you don't really care about your cell current density. What you care about is your CapEx, your OpEx of course and your land use. And some of the study -- I mean, Gareth, you did a bit of looking into this, really comparing -- we talked about that 120 square meters for the electrolyzer module in terms of how much hydrogen are you producing per square meter of land, right? And I think we're coming at pretty favorable in terms of our production density, I think. Yes.

Yes, I think we were comparable or better than alkaline, from memory. I think PEM intrinsically is just a little bit higher density. But I think we're a million miles away from some PEM manufacturers depending on which ones you looked at that. And it all comes down to how they do their balance plant as well. So if you look at a smaller scale, they're quite comparable. But as you scale up, if that company hasn't done the vertical scaling of their balance of plant, then their footprint explodes. It doesn't stay the same. You get a much more linear increase in that where actually, you take that off quite a bit by doing the design reference that we've used.

And on a follow-up question was coming in on the pressurization of the stacks. Will you also continue to develop atmospheric non-pressurized SOEC modules? Or will everything in the future probably be designed to be pressurized at that sort of 2 bar stack level that then goes to 30 bar?

Well, look, I mean, that's the direction of travel that we're looking at, at the moment. But again, we're a licensing company, right? So if a company wants to make atmospheric SOEC, then they're absolutely welcome to do that, and we would sell them a license to do that. Absolutely no problem at all. It's just for our studies, we see -- in terms of industrial applications, industrial integration, we see a real advantage of being able to have a reference design of the technology, which can operate in a pressurized manner, but it doesn't have to.

That makes sense. And Gareth, I mean, I wonder for the hydrogen hubs if nuclear has been involved in your hub. And if it has, I mean, nuclear obviously would have some key -- I mean, clearly, waste here is present quite a lot, some key attributes to be associated with -- if you were to build the plant really close, that would make a lot of sense. So have you seen a lot of studies that have started to take based on nuclear? Or it's still very early for nuclear and solid oxide to be paired together?

I've seen quite -- roughly a couple of studies now, and we've done some studies ourselves. And we're also actually doing a study at the moment looking at -- based on Ceres' technology actually for integrating to our own nuclear power plant design. So yes, like you said, you have steam available and you have electricity available. So both of those things are the key inputs. In terms of the -- in the hub, no, there's, I think, one nuclear power plant in that part of the U.S., meaning there's huge amounts of hydro, so the Pacific Northwest has that. But other plants in the U.S., what a lot of people don't realize is that nuclear power plants are actually -- most large thermal-generating plants actually use hydrogen on site to actually cool their generator units, [ the cool stat ]. And some of the plants in the U.S. are actually importing liquid hydrogen in large trailers 2 or 3 times a week just to cool the back of the power plant with hydrogen. So while nuclear power plants are [ widely favored ] using hydrogen on site, so there's no general issue with integrating this with it like you would with any other industrial facility. And it's a good partner because it can run all the time. Basically, they run with 95% to 97% capacity factors on nuclear. So yes, I think you can look at use case. I think if you can get -- if you get to the right ones or the right price, so a lot of the fully depreciated plants in the U.S. have got their very low constant power prices. They will be great to integrate with to develop hydrogen for industry, I think.

And Gareth, I'm just thinking now about -- obviously, we've spoken about the different technology classes, spoken about the scale-up and the opportunity for solid oxide. But what have been the main sort of challenges on the engineering side with solid oxide as a technology class? So there's still hurdles that need to be jumped over. A question here just asking what are the negatives. Can you give me the downside, right? What could potentially be the weakness for the technology? We know the weakness is -- for alkaline is potentially a lack of flexibility, also very big unit size. It's hard to move around, potentially hard to service. We know that PEM can be expensive. Iridium is present and maybe serviceability might also be a challenge and recoverability of the Iridium. What kind of stuff should we be thinking about on the electrolyzer side that you think could be a challenge for solid oxide?

One of the challenges for solid oxide is the life of which stack lasts for and some of the stuff we've been working with Ceres is really helping and we hope extend that out, I think, quite significantly. And that adds cost because you're probably changing it out maybe once, twice extra in the life of the plant versus an alkaline or PEM stack. But that's why you -- when you do the overall analysis, you still end up with 10% to 20% lower cost hydrogen even when you account for that, those change outs. So Jon, I don't know if you want to comment on the stack life and those other things maybe.

Yes, I think you're right. So certainly, solid oxide in terms of its technology class, it's been seen as quite expensive from a CapEx point of view because of the materials issues and those sort of things and robustness, not just in terms of just general durability but also in terms of cycling. People say, oh, you can't only cycle it. You can't load follow, et cetera, et cetera. Now not all those things are true actually. So we can certainly load follow but ramping up -- ramp it -- so the example that is normally used is you've got a wind farm and you're using curtailed wind and you want it to ramp up when the wind blows and then shut down. And how are you going to ramp the temperature up and down in time? And that's sometimes one of the questions, right? But actually, for solid oxide, I think you would keep it at temperature. So you warm it up and you leave it at a hot idle, we call it, and it's just ticking over. And then when power is available, it operates and then it ramps down again. But the point being really that, I think, Gareth alluded to the point earlier that, actually, if you got an asset, you've got an electrolyzer asset you spent many millions of dollars or euros on, you want to be sweating that asset massively. And that's really where you'd use it for an industrial application. You're really getting the most out of it in that regard. Now of course, I suppose, we at Ceres are trying to address some of those issues in terms of our technology to try to approach the cost, the CapEx cost by the lower temperature point and also the robustness to -- so it's not a purely ceramic high cost stack, which is quite sensitive to thermal shocks. We've got a steel architecture, which is actually quite robust to client trips and shutdowns and emergency stops and those sort of things. So that's kind of what -- the angle that we're trying to take is that, actually, we're trying to address some of those downsides of solid oxide through our particular technology basis.

Yes. And I suppose I'd also add from an industry point of view, PEM has been developing. So the stack life has improved for the -- you're still probably on that journey in terms of stack life of solid oxide. And you guys are on a lower temperature level. So would that potentially help as well on stack life or is it similar across the different competitors that you see in the market?

It does. And the other point I'll make, it's plant life really you're worried about, not stack life stale. So clearly, stack life is important but carry out some fairly good approaches in terms of your overall plant, making sure that you're -- the steam still coming in is of a certain quality, really trying to reduce the ways poisons can get into the stack. And it's something we're looking at, is, actually, do you need air flow, do you need air flow at all because that's definitely the way that poisons can be introduced into stacks. So thoughtful plant and system design can really -- anything -- a bit like a car really. You wouldn't run a car without a fuel filter or an air filter and the same with stacks. You've got to select materials correctly, get the right filtration in place, get the right operating strategies. And then you can actually see some really significant overall plant lives.

Just coming up on the hour, so last question to wrap it up, a question coming in on the largest electrolyzer stacks for solid oxide out there at the moment. So it's a 4-megawatt system maybe from Bloom Energy. When, do you believe, Ceres could be in place to deploy their first 10-megawatt unit with a partner or 20-megawatt system? When do you think that would be feasible if a partner came knocking on the door today to ask?

Right. Well, great question. And I think that's really what we've been working towards. So the work that we've been doing with [ CAF ], we've got that reference design available now. So that 8.6-megawatt design is now ready to go into detailed design. We've got stack manufacturing of that capacity available. And we're just going to start building the first stack array module, prototype stack array module this year. So next steps are exactly that, finding a commercial pilot project where we can site that 10-megawatt scale commercial pilot. Lots of conversations going on at the moment, haven't got a definitive answer to that question just yet, but that's definitely our next step.

Brilliant. Well a great place to end. Thank you both, Gareth and Jon, for your time today and for the audience participation for the very insightful questions. And I'll now pass it back to Paul.

That's great. Chris, Gareth, Jon, Elizabeth, thank you once again for updating attendees this afternoon. Ladies and gentlemen, could I please ask you not to close this session? We'll now automatically redirect you for the opportunity to provide your feedback in order that the company can really better understand your views and expectations. This will only take a few moments to complete, but I'm sure it will be greatly valued by the company. On behalf of Ceres Power Holdings plc, would like to thank you very much for your time this afternoon. Good afternoon to you all.