Speed to Power vs Net Zero: The Data Center Dilemma - Clarke Energy

Imagine opening a restaurant in a busy city, but the electricity company tells you the kitchen won't be connected for another three years. What a nightmare. Customers still arrive tomorrow, so what do you do? You rent generators, park them behind the building, and start cooking anyway. That's increasingly what the data center boom looks like. Every data center being built right now needs power immediately, not when the grid connection finally arrives.

That gap has created a market, and the technology filling it isn't new. It's the old school gas engine. Modular, fast to deploy, fast to start.

The same reciprocating engines that have powered factories, cars, and hospitals for decades. What has now changed is the scale.

Five years ago, a twenty megawatt distributed generation product was considered large.

Now Clark Energy is working on two hundred megawatt data center installations in the US and a four hundred and fifty megawatt peaking station outside London where banks of gas engines sit idle until wind drops and the grid calls for backup. Houston, we have a problem. But speed to power and their carbonization don't always pull in the same direction.

As my guest today puts it, the engineering department and the sustainability department don't always speak the same language.

Alex Marshall is VP of business development at Claric Energy with three decades across biogas, combined heat and power, and flexible grid support.

This episode isn't about whether gas has a future. It's about who controls the power when the grid can't. I'm Alejandro Adiego filling in for Ed Porter. Welcome back to transmission.

Alex, thank you very much for joining us today to Transmission, our podcast. Can you first start by introducing yourself, what your role is within the energy industry, and what, Clark Energy does. Yeah.

Thank you, Alex. So, yeah, I'm Alex Marshall. I've been with Clark Energy eighteen years, originally a biologist and moved in through, a strange route in the waste management sector, developing waste treatment technologies which had a biogas output, and that biogas output is used in gas engines. I worked my way through Clark Energy many years, relocating here to the US, almost four years ago now. I am a vice president of the Cogent World Coalition, and I sit on the world I assist on the council of the World Biogas Association. So Clark Energy, we are a distributed energy specialist, part of RELCO.

RELCO, is what was coal energy, and we focus on resilient, distributed energy solutions. So from a Clark Energy perspective, thirty years experience, we've seen the industry of, you know, local power ride a series of waves, whether that is the landfill sector, the biogas sector, combined heat and power, more recently flexible generation for grid support, in Australia remote power for mining, and then as we're here, obviously the big exciting topic is data centers.

We've been working on data centers for ten years now, and we're starting to see some really exciting projects coming through.

And when you work with new clients, do you usually focus on some technologies?

Do you look at all of the technology spectrum of power generation.

So our core business, we are the largest distributor of INEOS Janbakke gas engines, which is an Austrian gas engine. That kind of forms the nucleus of most of the projects we work on. However, you know, one of my roles in Clark Energy is to support the transition to new technologies for us. So what we do if you're talking from a client perspective, we ask the client: what what are your energy needs?

What's your, you know, electrical demand, your cooling or thermal demand? And then we would map on a solution to to match that requirement. If we start to build added levels of complexity, we can start to then look at, say, to the customer: well, you know, the energy trilemma: what's important to you? Is it cost?

Is it carbon emissions? Or is it resilience or availability?

And you can develop a, hybrid solution with with that trilemma, that triangle, to give the optimal solution for the customer. So for example, we could integrate batteries, or we could integrate on-site generation of hydrogen. We could do, a microgrid control system. We could put in solar, and and and bolt these together based upon what the customer requires.

And that's that's we have we have some specific suppliers, then for other scopes of work, we are we are agnostic. We'll go in. We're a life of asset provider, so an EPC, engineer, procure and construct, but then with maintenance. So we'll give an interest in the in the facility for maybe ten, fifteen years into the future.

You have been at the front of hyperscales data centers, as you mentioned, right now, and recently in the US as well.

Has there been a sheer moment when you realized the boom that was coming with with a specific project or a certain year? And what did it translate to in your daily operations or your strategy for Clarke Energy?

Yeah. That's a good question. So from a from a marketing perspective, we've been watching this market for many years, and to be honest it has been quite frustrating. Ten years ago we identified this as an opportunity, but the data centers were comfortable with grid electricity. They wanted energy generation as a backup, not for baseload power.

So we saw early projects there, projects going into Ireland small enterprise data centers one, two megawatts where it was next to an office they had a requirement for heat, and that heat would be fed to the office.

We did some early rail operating centers in the UK, and then one of the most exciting projects was at Citibank where they it was Green Investment Bank financed. This installation again on the small scale they were converting the heat to cooling via absorption chilling so combined cooling and power that demonstrate the carbon efficiency of that technology. Now we started to see, you know, historically for us we were working on one, two megawatt projects across the market maybe ten was an unusual one. A twenty or a fifty megawatt project was a unicorn project.

Now we saw in Ireland we had a sixty megawatt project which we kind of shied back from a little bit thinking, you know, is this too big? Is it too much risk? Now that project is small for us. We're starting to see a rolling rolling rolling kind of pipeline.

From our experience on that project and a further ninety megawatt project in Ireland, the we were in the US, but still the US wasn't talking when I arrived, really self generation of power data centers. Yes. On the data center side, where we scored a two hundred megawatt project recently, and wow, this is this this kind of shows you that the the world is on a an inflection curve for the demand for the power. So I would say once we saw that market move out of the Irish kind of example projects into the US and saw the growth rate here, that's when it's like, this is really really big here, but also globally. We operate in Australia and India, and this is this wave is following other countries as well.

What is the biggest project you are working on right now in terms of capacity?

It's not a data center project. It's a it's a peaking station project. We're just finalizing that. It's a four fifty megawatt station outside London. Very similar technology to to that for data centers, and that's also one of the opportunities for data centers. The peaking stations sit there and provide power when there's a fall off in wind or a fall off in solar power, and obviously the baseload power can no longer cape. So those units come on, signal from the grid saying price is high, meaning no supply is low, and they turn on.

I see. And what is the advantage that these peaking plants can provide compared to batteries right now? Because we're seeing very big batteries being built. Yeah. What is their advantages?

So batteries typically want to duration output in the the Dunkelflout, the low wind doldrums, the low the low energy doldrums of the winter, when you have no no sun or wind for a week, those batteries simply just don't, you know, don't fit that gap. And batteries don't generate electricity, they store electricity. So you gas engines and gas peaking stations will fill that gap when batteries can't do that. And there are don't get me wrong. Batteries and engines work very well. They're complementary technologies because batteries can ramp ramp very, very quickly. Engines can ramp quickly but not quite as efficiently as so you can marry those two technologies in in parallel.

So so that's on the on the the peaking station side. Incidentally, early ninety megawatt project we did in Ireland for a data center was forty megawatts baseload for that facility, forty megawatts flexible energy in support of the grid, I. E. The data center no longer being selfish, just thinking about their own operation, but actually providing residual grid support and then some redundancy in there to make full full availability of the assets.

Okay. I see. And when a customer comes here in the US and asks for data center on-site generation, what is the typical setup of technologies that you have for data centers here in the US?

So what we see so we we can offer the full suite of microgrid technologies, but in reality data centers, you know, we're looking at primarily, you know, historically back at diesel. We're looking at baseload gas or bridging power gas, and then energy storage. And if the data center is more kind of energy conscious, environmental conscious, we'll see the deployment of combined cooling and biotechnology as well.

You also mentioned that you offer, combined heated power solutions. What are the type of customers that go for that on-site, solution rather than just pure power generation?

Well, that that's a that's a that's a good question. And, you know, the the early reference project in London proved the concept. But in reality, data centers, you know, they're concerned about, speed to power. They're not necessarily concerned at the moment about efficiency. I think we'll see those customers go through a, you know, a, a learning process themselves, and they realize that maximizing the value of every electron of the, the fuel is important.

So if you're in a bridging solution, you know, if it's three to five years operation, those solutions are less likely to see combined cooling power because it means additional capital expenditure. So the facility still has to pay off its capital expenditure within that time frame. If they're talking baseload over the life of the asset, then that becomes a lot more attractive and logical, and there's a greater degree of time to to to pay off that investment.

And from the projects that you're seeing now, what are the deciding factors for those developers to go for on-site generation rather grid connection?

Well, it's simple. A, availability of electrons in the local network or b, the ability to bring that level of power to that facility in a given point of time. So it's constraint.

Has there ever been an occasion where you said to potential customers, hey, this doesn't make sense, you should actually go to the grid rather than looking for on-site generation?

Well, normally the customer will know that in advance, but yes, if you've got very low electricity prices, you know, typically where there's a nuclear power station which has got surplus energy, but even those power stations are being, you know, hoovered up their capacity. So that's it's changing.

Okay. And what would you say are the biggest obstacles right now, generally, for deployment of on-site generation?

Location of the site, whether or not there's physical space to do that, the cost of power locally, the willingness of the data center operator to self generate power, because data centers don't it's not their core business. These these these are sort of factors I think.

With Did you have the need to say no to some customers due to supply chain reasons, a lot of backlog?

Or No. We haven't said no to any customers, but, yes, lead times are expanding. Yeah. We still have a attractive lead time compared to some of the options right now, but, yes, you know, demand is high and it's low suppliers market at the moment.

Okay. And within the US, are you seeing a concentration of a lot of your price in some regions or markets?

What we're finding is more along the less populated central states of the US, the likes of Utah, maybe more agricultural states like Ohio. These these areas which have got lots of space, relatively cheap power, that's where we're finding a lot of the newer installations going in, where, you know, places like Virginia are hugely crowded, overcapacity, and, you know, you've got some challenges in those areas still.

You spend a lot of time publicly making the case for a net zero pathway in the future, And in that sense, what's the most defensible role for modern gas engines for the next ten to fifteen years, especially taking into account the transition to a net zero pathway?

Yeah. That's a very good question, and there's probably a series of steps that need to be taken. So even using natural gas, if you use combined heat and power configuration Yes. So you use the electricity and the heat or you convert the heat to cooling, you can achieve a system efficiency of ninety percent.

In almost all locations in the world, that still has lower carbon emissions than the grid electricity. So before we even transition to fully renewable fuels, the ability of CHP to deliver carbon savings is tangible and immediate. Yeah. The next level is time efficiency.

So as mentioned before, the grid the the grid is on there's more and more wind and solar on the grid, makes the grid unstable. The the data center so the data center operators and gas engine technology can support the grid in times of the renewable true renewables being unable to supply. That's the next phase. Then if we work up that, can also, you know, on a on a more advanced level, could add carbon capture.

Post combustion carbon capture is proven technology, widely deployed, and that CO2 is a valuable offtake product that could be could be taken. Then you talk about renewable fuels. So the gas engines deployed today can operate directly on RNG renewable natural gas molecularly the same as as natural gas and a renewable fuel or you can convert to hydrogen. And the engines are hydrogen ready and can operate anywhere from zero percent hydrogen to one hundred percent hydrogen with a balanced natural gas.

So you can decarbonize at, you know, intervals depending upon what's available locally, what fuel supply is there, and the other component is gas engines can form the backbone of microgrids, whether they be renewable fuel or natural gas fuel, integrating the solar battery storage systems and providing that firm power when those two systems aren't aren't fully operational.

And usually when clients come to you, you said the biggest reason is speed to power.

Do they ever come to you for cost savings purposes? Have you seen cases where it was way cheaper for them to have on-site generation than connecting to the grid?

That's a very good question. So in the commercial industrial space, that is almost essentially the entire, you know, objective of combining a power plant. Rather than generate electricity or take electricity from the grid, generate heat from natural gas and use that on-site, you put a CHP unit in to do those factors and the aim is typically to reduce cost. So you'll see a return on investment within, you know, one to two years of operation in many instances.

And usually when you deploy a solution you take care of operations and maintenance as well?

Yeah, that's where we come from.

Okay.

So we're a life of assets energy solutions company.

Yeah.

So rather than just engineering procurement and construction, we had maintenance to take the the life of the asset into consideration.

Do you see a future where a coupling of renewables plus batteries will fully replace thermal plants or you don't think it will come in the long term?

I think they'll certainly eat into the until lunch of the thermal plants. The difficulty is, you know, I think you're from Spain, some of the challenges on inertia on the grid is, you know, recently surfaced. So, clearly renewables are good, but they need to also provide grid stability. So there's still the role of rotating equipment, I. E. Thermal generation, in that in that platform.

You mentioned firm generation for grid stability in terms of voltage and frequency. But when we're thinking about on-site generation, which is an island, those services are not that much needed in terms of keeping the frequency in the grid. It's micro island.

In that sense, if the com the thermal plant is not offering that service to the on-site loads, do you still think that there's an advantage compared to batteries coupled with renewables?

So islands by the definition are inherently unstable, so the generation assets have to handle with those loads. So AI is notoriously peaky, and you've got very large loads coming on and dropping off. That's where, you know, using engines in parallel with batteries is very good, because you can have those two things working in parallel. We've done many projects globally where, you know, for example in Nigeria we have a plastics extrusion site. The the loading rates of those generators, because they can't connect to the grid, has to has to absorb those those challenges.

Those issues are serious, but they're normally kind of addressed through some sort of energy storage technology. So there's kind of synergies with those. So, like, ultracapacitors or, or, you know, battery energy storage systems. And these these kind of hybrids are being deployed in in the data center space.

Okay. And now that you have experience working in this distributed energy space in the in Europe, in the UK, and also in the US, are there any main differences that you see in in the way the key stakeholders, the way they look for into these projects, what value they see in them between the geographies?

Yeah. I think I think sustainability is higher up the European agenda. That's gonna that's gonna have got a double, you know, it's a double edged sword. Yes.

There's more kind of potential resilience from renewables but also it's pushing the prices up in Europe, which I think people are starting to feel the pain. US maybe a little bit more gung ho on the fossil fuels and they would benefit, I believe, from more energy efficient thinking because these fossil fuels are finite as well. You know, if we're using them, let's use them efficiently. So I think those those are probably the two the two main considerations.

You know, data centers in Europe for almost certainly if they're gas fueled, they're gonna be CHP from the outset, whereas in the US the consideration is different.

Before we jump onto the final section, is there anything that should have asked you that would be interesting for our audience?

So one of the points we can perhaps mention is the, the potential for the production tax credit. Combined heat and power, under the previous inflation reduction act and the previous government's policies fell between the cracks because it wasn't a fully renewable technology. It's an energy efficient technology.

The current administration is more keen on natural gas, but still some of these support mechanisms aren't favored. One of the ones that's tabled at the moment is the 45Y production tax credit, which was, which had a note from the the treasury recently that they they approved of this, it hasn't been ratified by the federal government. What this would mean is it would improve energy efficiency and speed to market for combined cooling and power in the data center space. So what would happen in essence is potentially at the outset of a project you could monetise the production tax credit for the thermal value of the energy over the life of the installation at year one by giving those credits to a bank.

They would monetise that, give the developer a, a sum of money, which could then be used to offset the cost of the absorption chilling. That would massively help the path to improved efficiency, reduced carbon emissions, and also, you know, better use of the domestic fuel reserves, which shouldn't be burnt off too quickly.

Yeah. And what technology specifically can access this fortify? Why credit?

Is it combined heated power, but all types of thermal plants or what's thermal This is combined heating power, and it's primarily aimed at larger thermal plants.

Okay.

Are you using natural gas? Also, for example, coal?

That's a good question. I don't know the answer to that. Certainly natural gas by, I guess, hydrogen, I don't know the answer on coal.

Okay. Perfect. Thank you. And whenever we have talked about customers putting first speed to power, are the customers looking at costs? For the customers looking at emissions, and mostly that could be Europe, are there any customers in the US that are really taking care of reducing their emission levels? And if that's the case, what type of customers are those usually?

Well, talking about the very big names in the market, I won't mention any clearly, but you'll have these very large corporations and they'll have, you know, an engineering department tasked with, know, putting an installation to the ground very quickly, and then you have the sustainability department talking about the importance of reduced emissions. These two functions don't really talk to each other. The it's it's the the kind of the business need functions which are taking priority, clearly, and I think they would benefit from having a probably a greater integration and a greater say.

Some companies, some, you know, a lot of people talk about their carbon goals, their decarbonization goals, their pathway to net zero. I think realistically a lot of these are very high. I don't think many people will actually achieve that. There will be one or two that do that, But I suspect it won't be the entire market.

Okay. And just jumping onto another topic, which I found out of curiosity, very interesting is, what is the actual business model of Clark Energy in itself? Is it having their final customers pay for all of the CapEx of the new plants and then having revenues come from the service operating and maintaining those plants or what exactly is the business model?

Simply engineering, installation, maintenance.

We supply a piece of equipment, we charge we have some margin on selling the equipment, and then we have a long term service contract which we support availability and we receive regular payments. So those two things form our secret sauce: it's the upfront installation, a long term service tail.

And when you get access to those forty five y credits, is it you that well, Clark Energy, the ones that are managing the value of those tax credits set?

No. So it's the final customer?

That that would be the final customer. Their responsibility. Would be their benefit too to install that. We we're not a developer.

We typically don't own the assets. That would be the customer. Where we where we kind of differentiate ourselves is a quality installation backed up by reliable after sales support. There's a lot of peaky people at the moment putting equipment in, and we find this on each market.

There's a series of waves. The early first movers, they just want low cost, speed. They realize that, you know, on-site power is about availability, and availability needs to be done by quality engineering at the outset, and then trained reputable service technicians at the tail. And there's already, you know, not enough of those individuals in the US in particular, but globally, and we're one of the few companies that can bring that to market and do that successfully.

So there's going be a learning curve, I think.

And can you tell us about a specific project that you've been very proud of, being part of it, be it for an innovative solution, an innovative setup, or a very tough region with high prices whatsoever? Can you tell us about a specific project? Yeah.

Well, there's there's there's several of them.

We've we've worked on projects in in Ireland which recover the engine exhaust and convert that to beverage grade c o two, I. E. The carbonation in drinks, that's quite important. You know, there's some major, biogas upgrading facilities in Indiana where, you know, we provide renewable electricity, renewable heat, and renewable natural gas for injection into the network. I think that's really good.

The early combined cooling and power project with Citibank that we did ten years ago, that's demonstrated that that technology in the market. So there's many many kind of innovative application specific projects we've done. We've done compressed natural gas projects in remote areas of Nigeria where they've historically relying on diesel generation as baseload power, and then we're also doing hybrid installations. We have the hybrid engine battery project one in Romania, one in Connecticut, and one in Australia proving the concept of these technologies in different countries, not just in a very narrow one geographical mindset.

And usually in all of these countries that you mentioned, is it industrial customers usually, also data centres?

What type of use cases are we in?

Yeah.

Ranges, you know, data centers, commercial industrial facilities, oil and gas facilities one in Nigeria, for Nigeria is a a a manufacturer.

Okay.

Great. And in Romania?

Romania is a grid support system. Yeah. Perhaps also interesting for the data center market is we finally in Eastern Europe, have lots of legacy Soviet, district energy networks. These are coming to the end of life. You know, they're historically large large old turbine, inefficient turbine fuel power generation. We're starting to refurbish those pieces of equipment, but data centers, if they had joined up syncing, could link next to these networks and and use the surplus heat from the from the on-site power generation into those networks to to offset some of the local loads.

That's actually a very interesting point. So all around Eastern Europe, there are these facilities that are outdated old, not used anymore, and you can refurbish them and bring some of the customer load to those locations to restart the plants, actually make use of them. Is that correct?

So we are doing refurbishment of the district energy schemes. At the moment, there's no data center associated, so we're refurbishing with new assets or building new projects in Greece.

However, if a data center operator worked with a district energy developer, there would be a natural off taking for the heat from the data center that could be used and used more efficiently.

Okay, I see. Well, thank you for sharing that very interesting point. We're jumping now to the final part.

First question is whether you would like to pluck or promote anything to our audience. This is the moment for it right now.

The most important thing from our perspective is working with a quality supplier like Clark Energy, able to give, you know, long term availability and performance for a project. So if you have a project that needs reliable power, don't just think installation, think about installation and after sales support and that's where we can help.

And now the final question that we all ask to all of our guests is what is your confirmed view on the energy industry that many people would not share with you?

Yeah. So I come from the organic waste sector and from the biogas sector. Anaerobic digestion of organic waste properly deployed globally could contribute to around about eleven percent reduction in carbon emissions.

From a US perspective, yes, there is biogas sector, a lot in the the cow manure, a lot for landfills, but there's a huge under under penetration from organic waste treatment.

The the US, if it started to better connect domestic households to waste treatment networks, you know, there's in some areas fifty percent of houses are on septic systems, if food waste was better collected and not dumped in landfills, this organic waste can be converted to biogas, injected into the gas network, and with support from the data center sector, the ones that identified this as a fuel they were willing to pay a small premium for could stimulate that market and have, you know, big knock on benefits one for decarbonisation of the data centres or defossilization, I should say, as opposed to decarbonization, and it would support reduction in emissions from organic waste. You know, think septic systems account for something like seventy five percent of the emissions from, the US waste waste treatment sector, and there's not a lot of drive to do that.

And have you ever studied any regions within the US with higher potential for these with this for these solutions?

So I'm I'm I'm from Old England, living in New England. Yeah. In in Boston and around those areas, surprisingly, the penetration of septic systems is particularly high. Yeah.

So in in in East Coast US, I think there's, you know, a huge opportunity to do more. Most biogas projects are centered around California, which is driving the market, or the Midwest where you have the agriculture. But I've I've I think that the East Coast has huge under penetration on the food and human waste side, and then if you look to other parts, I've been to sites in California that are still uncovered lagoons of organic waste Yes. Which are emitting methane literally bubbling straight to atmosphere.

Capture that. Use it to decarbonize the gas grid further, and there's huge potential.

And for our audience to really understand in simple terms what it would translate to an actual work to be done, you will need to build a biogas plant generating power, but at the same time, you will need to build all of the infrastructure behind to process that waste and transport it. Is that correct?

Depends on the sector.

On those those uncovered lagoons Yes.

You pretty much need to just cap them, capture the gas, and and convert it. In the waste sector and the household sector there needs to be support for households to connect to a centralised sewage treatment network. And then from the organic waste side people need to collect and treat food waste separately from the rest of their bin, I. E.

Separate collections or separate collection stations that have localized anaerobic digestion systems that have been deployed widely in other countries. You mentioned the small price premium for those solutions. What are we talking about? Do you have any number?

The potential figures off my head, but I think it's in the region of twenty to thirty percent higher than natural gas. Okay.

Still significant, but nonetheless Well, I'm sure that some of the developer actual loads will think about this solution. This might open some doors.

I I hope so.

So, Alex, thank you very much for joining us today. It was a pleasure to have you here with us, and thank you for sharing such an interesting Contouring view.

Thank you. Much appreciated.