This is Tom Rowlands Reese and you're listening to Switched on the BNF podcast.

Those of us who work on the energy transition often find ourselves studying technologies that seem like science fiction.

Harnessing the sun's power to fill your car, for instance, or using the wind to run data centers.

Of course, both of these examples not only exist, they're also cheap and practical enough for widespread adoption.

Direct air capture is another one of those technologies.

Can we really invent a machine to pull carbon out of the atmosphere, putting a lid on humanity's emissions and getting the world on track for net zero?

The good news is that technology absolutely exists.

The less good news is it's really really expensive.

Today.

Direct air capture or DAK costs around nine hundred dollars on average for capturing a metric ton of carbon dioxide, and the process is extremely energy intensive.

Given that cheaper means of carbon and removal exist, including a relatively simple one planting trees, it'd be fair to ask whether the direct air capture is really a viable decarbonization option, and if so, what'll it take to get us there?

Today I'm joined by a member of bnef's Stemable Materials team associate Brenna Casey, and we discuss findings from her note out of Thin Air the cost of scaling direct air Capture, which BNAF clients can find at bn EF go on the Bloomberg terminal or on bnef dot com.

All right, let's get talking about the outlook for the cost of DAK with Brenner.

Welcome to the podcast, Brenna Casey.

Thanks for having me.

Tom Rolands Rees.

I mean, there's a reason that I'm taking us away from just using first names here is because I'm going to be a little bit unfriendly today to start with, and for people listening.

I did warn Brenner about this beforehand.

We went for a coffee couple of days ago and I said, you know, we're doing this podcast.

You published a piece of research about the future cost outlook for dak AKA direct air capture.

We'll get into that in a moment, but before we even do that, I'm going to question the very existence of why we are even covering DAK and why anyone is even trying with DAK so Brenna Casey direct Air Capture.

So we can get into the technicalities, but is basically taking common dioxide out of the air and storing it in the ground with a view to reducing emissions primarily.

And I suppose the reason I say why would we question its whole existence is we're in a world right now where people are not talking about net zero anymore.

People are even questioning the legitimacy of electric vehicles, of renewables.

These are things that can compete with their alternatives even when we don't consider environmental factors.

Whereas DAK only makes sense in a world where we're trying to get to net zero because it just serves no other purpose.

And so, given the way that the pendulum has swung, surely that's curtains for direct air capture as a promising new technology.

What would you say to that?

Rather, let me say somewhat downbeat assessment of the state of flavor direct air capture.

Yeah, so I guess from a system's approach, why we need direct our capture, and then I'll get into I guess more theoretical, kind of upbeat side of why we need DAK.

Basically, CCS can only abate so many emissions, right, So if you're.

Looking at ccs's common capture and storage.

Yeah, exactly.

So just for an example to put this into context, the cement industry or the petrochemical industry, they have process emissions just inherent to the chemical process itself, which can only be abated so much by fuel switching or electrification things like that.

So you need ccs, But even ccs can only abate conservatively ninety percent of emissions.

There's things like oxy fuel combustion or other newer technologies which can up that emissions abatement ninety five to ninty nine percent, but there will still be these unaddressed emissions, right, which you need DAK for.

Right, So it would be to cover that last ten percent of emissions again that other sources are not.

But the cynic in me would say that this sounds very very twenty twenty three, twenty twenty four.

Yes, you know, and particularly I realize I'm looking at this from a US centric lens, but the US does set the tone for a lot of what is happening elsewhere.

It's been a change of administration, a lot of the impetus and support has changed, and in particular the Inflation Reduction Act, a lot of that has been unraveled with the big Beautiful Bill Act.

And so in that scenario where there are things that maybe have an emissions impact but also have other benefits, like you know, potentially solar is cheaper than fossil fuel alterns.

In those scenarios, why would we think about DAK Because what you were just saying is like, well, we can't reduce emissions solely with these technologies, and you know you need DAK to fill in the gap.

But if you're, say, working at a company that is behind direct air capture, what is your value proposition in a world where there is not a sufficient incentive coming from anywhere to eliminate emissions completely if the premise of the point of DAK is to eliminate those emissions that can't be eliminated through other means, right.

I mean, it's an interesting question because in the United States, you're right, there isn't this third party, which is normally the government saying by DAK.

Yeah, it's almost always an added cost, similar to the way CCS is almost always an added cost.

It's a waste management technique.

You know, it's not commercially viable, and any mean, the only way DAK is going to scale or be adopted by anybody really is one a technological step change which will help costs come down because they are prohibitively expensive.

And that's what we'll get to them.

Yeah.

The first, I think most important thing is obviously demand, and you're not going to have demand right now unless you are a Microsoft or a Shopify or Stripe with this tremendous good will and they're buying credit.

So what you need on top of that is compliance markets because right now all DAK is traded either B to B business business or on these voluntary markets.

And so you need to see this transition from DAK into again a compliance market.

And we actually just saw kind of the beginnings of this with the UK.

The government over there said hey, we're going to bake CDR into our.

Cap Entrade program.

CDR stands for carbon dioxide removal, Okay, so DAK DAK, yeahrect Derek Apsher, Yeah, there are stipulations with that, the DAK plant has to be in the UK, things like that, but it's a first move, right And then again there are whispers of the EU potentially baking in CDR into their cap and trade program by twenty thirty maybe twenty thirty five.

But these are the sorts of things that is going to help DAK commercialize.

Got it?

And I think you said something there that before we got onto compliance markets.

I can see what your point is that that is, like, I suppose the holy grail for DAK is for there to be meaningful compliance markets around the world, and DAK is an option in there.

But maybe you answered my question is, so why this is not completely dead in the water, is that the voluntary markets still matter?

Regin?

You know, just because the US federal government is swinging one way doesn't necessarily mean that all of the companies in the US or worldwide is swinging that direction.

And is that what is the current market and the current sort of hope in the near term for DAK is their use in voluntary carbon markets.

Yeah, I mean I think we can we can take this and look back to what happened with solar.

So Reagan cut solar, just cut it all, and then what do.

You mean he cut it all?

He just dropped funding for solar.

Okay, for what for solar?

Like research and development.

Exactly, Yeah, and subsidies, et cetera.

He ripped the solar panels off the lighthouse, and then what happened is Solar died in the United States and then moved to Europe because Germany was offering so many subsidies to kickstart the market there.

And then we saw the market move there, we saw the people move there, and then again then we saw, you know, through other sorts of policy schemes, we then saw solar moved to China.

And so this could be maybe a window, I would say, into what's happening right now.

It's just because there is this negative sentiment in the United States, and I guess let's address also.

The Trump administration has paused the subsidies for the dee dacos, but they're not fully revoked as of right now.

So a lot of companies are continuing, hesitant to say business as usual, but they are still continuing in the United States until that punning is totally gone.

We also did see the utilization credit for DAK came up to parody with storage.

You know, that's for eour purposes in a lot of ways, which got it backhanded for DAK.

But I think it's not entirely dead in the United States.

I think there's less support and maybe that's hurting the United States competitiveness right now.

But I think DAK will still continue in other pockets.

I'm a book market.

I mean what you're saying brings to mind.

So the head of b n EF, John Moore, loves to quote this British sports coach, so Clive Woodward, who predominantly famous for rugby, but he said success doesn't happen in straight lines.

So it's to say, you know, you don't start winning and then continue to win and then finish winning.

There are along that the road, there are bumps and setbacks.

And DAK is a climate technology in the most absolute sense.

And then that, like the climate question, is a lot more long term than any one single administration.

So I suppose what I'm hearing from you is that you know, firstly, it's not about one country, but it's also not about one moment in time, and that is why DAK is still there.

And I think when I spoke to you over coffee, I think it's clime Works have recently attracted significant funding.

They just raised one hundred and sixty two million in a series.

So that's not exactly consistent with the dead technology.

Yeah exactly.

I mean, you see, obviously what we just address, like we've seen political sentiment or political ambiguity, or just market sentiment in the United States maybe begin to falter as a response of what's happening politically.

But again, we just saw that climb Works deal.

We've seen JP Morgan with think it was fifty thousand tons of an off take agreement from one point five.

We saw United Airlines with a huge deal from Airlom long term off take agreement as well.

So the market there's still interest.

We are still seeing that interest.

So then I mean, let's move this on to the topic of your recent research, which is looking at at the costs, because I suppose you know, if we are saying this is a long term play, you know there are applications for it right now, but it's something that if things go according to plan, for DAK will become more and more relevant in the future.

And for that to happen, maybe the costs will come down.

And you can tell me.

I'm about to ask you about that in a second, but before we even talk about, you know, are the costs going to come down when it comes to like the way you've painted the role of DAK as part of a suite of emissions reductions options.

It's like the option of last resort.

It does the things that nothing else can do.

And I suppose a question I have is if it's the only option of last resort, then in a way it wouldn't be that sensitive to costs because it's like there's no other option and it's just prohibitively expensive.

But I'm assuming there might be other options.

So what is it competing against?

You know, what are the costs have to be competitive with for it to succeed?

Right, So I think you can take this too ways if you're looking at offsets just across the board.

Sure you're competing with deforestation or reforestation and clean cookstoves and those sorts of things.

There's biochar, there's bio energy with carbon capture of backs, there's direct ocean removals.

There's all sorts of things.

But not all of these are high integrity offsets, right right, we have do you to.

Just explain what we mean by high integrity offsets for people informidia?

Yeah, exactly.

So they have low permanence, meaning especially for direct ocean removal or a lot of those sorts of technologies, it's you don't know when the CU two will resurface or come back into the atmosphere.

You're sort of relying on some theory and you know you're part of a broader statistical picture.

Yeah, exactly.

So there's there's bad MRB or monitoring, reporting and verification of that particular offset, and we've seen that with just so many PR nightmares over the last couple of years, right, just thanks.

So aside from you know, obviously it needs to be close in costs, but what DACKS real magic source it has over the alternatives is it is the highest quality option.

Like a ton of CO two removed through DAK is better than a ton of CO two removed through something elsewhere there's a little bit more uncertainty about what it really is.

Yeah, I think.

I mean, we could have an entirely different conversation about biochar verst DACK or bex urst DAK.

But when you're looking at an engineered removal like DAK, and you are paying a premium for that removal, but paying that premium kind of removes this PR or even like regulatory risk that comes with buying some of these lower quality offsets.

And for that reason, we kind of see DAK as one of the gold standards of carbon removal, and I think a lot of the market sees that as well, and that's why you have Microsoft Stripe Shop Fight kind oft three.

Because I suppose if your incentive in the voluntary markets at least is as much to do with reputation, these alternatives leave you quite exposed to aixations of greenwashing, whereas DAK it's kind of unun No one can accuse you.

Of that exactly.

Yeah, let's move.

On to them talking about your research into costs.

And you know, we've talked about director capture, you know, as if it's a monolith.

And from what I understand from the work you've done is we've actually looked at three different technology pathways.

So maybe like we let's explain what those are and then we can start talking about their relative outlook for costs on all of them.

Right, So in this report, we have calcium maxide looping, which is pioneered by carbon engineering, a subsidiary of oxy.

Then we have solid sorbents, so that's mostly in this modeling solid amines, so you can think clim works on that front.

And then we have the more nascent novel suite of technologies, which is electrochemical.

So these guys are their glorified batteries, you can say, so they don't use any thermal energy, which mitigates a lot of the associated latent heat losses within the system, and so they have far lower energy requirements which could help them scale and realize far lower costs than the other two solutions.

Interesting, but that's the sort of the kind of the new kid on the block.

Yeah, that's parking lot stage, got a lab stage.

Nothing is just a commercial scale.

Yet I love how you've equated a lab and a parking lot.

All right, then, so I suppose let me just ask the question, what is the answer?

What did you find in terms of the kind of the cost trajectories of these technologies?

You know, how cost competitive are they relative to each other today?

And where do you think they'll be in the future.

Yeah, so I think maybe to put this into context even further, taking a very large step back back in I think it was twenty ten, twenty eleven, this report came out from was the APS American Physical Society, and they came out saying DAK will never reach below six hundred or five hundred dollars per ton.

This is again prohibitively expensive.

It's impossible.

Let's table this forever, and then I think it was twenty eighteen carbon Engineering David Keith came out and said, hey, listen, like, we have a technology and we can get this down anywhere from one hundred to two hundred dollars per ton.

This is commercially viable.

And ever since then, every kind of maybe incumbent DAK company and every company since then has been under this pressure now to align with that one hundred dollars per ton number to stay relevant in the marketplace.

And so we had this artificial kind of I was.

Going to say, because that baseline is just something that someone said once exactly when they were trying to get some pr exactly.

So we have seen the industry coalesce around this one hundred dollars per ton number.

That doesn't seem like a lot, you know, as in, it's not a million miles off from where certain compliance carbon markets are.

I mean, yeah, you know, and given the role we've described for DAC, it seems like if there was one hundred dollars, you know, we wouldn't have a climate crisis.

Would just be like pump in all the common dixide into the ground exactly exactly.

But again that's what The note was kind of built on what we've seen in the past is companies again, like as I just said, companies continue to announce these overly optimistic estimates to where their deck technology is going to go, and when you look at the thermodynamics of that process, it's impossible.

So what we found, if you're looking at technologies today, electro chemical is the most expensive, just because it is the most novel.

You're looking at costs around sixteen hundred dollars per ton of CO two solid sorbents today are going to be around that one thousand dollars per ton plus or minus range.

Those companies, again, it's the climb works the air means, it's remover that they're a zeolite company.

They're using zeolites instead of amines.

And then we have you know, carbon engineering calcium oxye looping that hovers it around you know, the four to maybe less than six hundred dollars per ton range.

The way all of those costs come down the curve are wildly different.

So Climorks Solid Sorbents is really is the only company today actually with the commercial scale project.

It's it's almost forty thousand ton per year facility.

It's not running at old capacity.

By any mean they have that going, then we have Carbon Engineering has a half a million ton plant set to come online by the end of this year.

I guess we'll start with solid sorbents.

So starting at this, let's say, for simplicity's sake, around one thousand dollars per ton range, we're likely to see moderate learning rates anywhere from four to seven percent, So you'll see a much more dramatic cost down than calcium oxide looping one because these guys are able to leverage modularity, so you have increased learnings from that.

But again, when you come to this twenty thirty five twenty fifty range, even with enhanced learnings learning by doing, you're still going to see solid sorbins maybe be.

One of the most expensive technologies at scale.

And that's because when you look at the capex breakdown, the only really novel component is the air contactor, which doesn't actually make up that significant of a portion of the capex, and so most of the component parts are off the shelf.

They're vacuum pumps or keat exchangers, things like that that which won't actually realize significant learnings because they're like reupholstered from other parts of industry, right, so.

It's a solid sorbent process, is built off of mature technologies and combination, and its biggest I guess advantage in terms of cost physics modularity is that what you say.

Yeah, you could say, that's probably one of the key things just because you have don't get I mean, don't get me wrong, because what we've seen climorks do is I mean they've released an announcement saying we have caught capex by fifty percent just because they've reconfigured the process designed to allow for more efficient like they've increased the kinetics of the reaction, so it happens quicker.

Amines themselves are.

Highly volatile, so that they degrade quite quickly in the presence of impurities, meaning they have to be replaced quite often, and they're actually really expensive, so that kicks up the oppex costs pretty significantly.

What they're doing is making that sortment itself far more efficient, so you can you can realize learnings, and people are still realizing learnings.

But at the end of the day, there is this bottom line, this thermodynamic limitation with the SORBM in itself that you just can't really, So.

That's sorting under, right, I'm curious to know, and I suppose this applies to the other technology pathways as well.

And what fraction of the cost of DACK is the capex and what fraction of it is the opex and in particular the energy consumed.

I think the biggest bottleneck for director capture, and probably one of the most scrutinized parts of director capture by the general public, is the energy consumption land use, but also energy consumption.

And if you're looking at you know, calcium oxye looping.

It it requires temperatures of up to nine hundred degree CE, right, And if you're looking at solid sorbents it requires you know, one to three hundred degree C.

But it's still a pretty.

Significant amount of the capex, especially for solid DAC approaches.

And this is energy that presumably can't be provided by electricity for those two processes.

And you know you mentioned electrochemical, which presumably it is powered by electricity.

Are these ones relying on fuels?

Right?

So l DAK or calcium oxyde looping is reliant right now on natural gas.

Right.

Oxy has said that they're looking to potentially use an ecal signer.

Airloom is working on creating an ecal signer.

They require high temperature heat as well.

Right now, the Climax plant in Iceland relies on geothermal, so that's twenty four seven clean.

But again that's at a very very small scale.

We'll come back to this question of energy maybe when we've talked about electric chemical because maybe I'm jumping the gun, but let's talk about the calcium oxide looping.

Tell us a bit about how that works and what its particular cost outlook is.

Right, So, calcium oxide looping it's around four to six hundred dollars today.

That tech is.

Looking at even lower learning rates than solid sorbents.

One.

That's because it's inherently it's large fully integrated.

It's half a million times already, so it's reached this natural economies of scale, so.

There's no economies of scale to be gotten out of it.

They and it can't be modulario.

There are components of it.

You could say that maybe the air contactors themselves are modular, whereas you know where you kind of flow the chemicals is integrated.

But fundamentally it's not going to achieve the same learnings as a functional modularity as electrochemical or solid sorbents well, again similar to STAK or solid a means those sorts of approaches.

Many of the component parts are off the shelf, so again mature technologies used in other parts of industry, and the air contactor, which is one of the only novel components, makes up a relatively small fraction of that capex, so you have very limited room to see costs come down that curve because each component will realize different learnings.

What we've done is we've aggregated those learnings into the four to five percent that we're going to see for liquid absorption or calcium oxy looping, and then the for five to seven that we're seeing for solid deck some of the.

Mature technologies calcium oxide looping and then solid absorption.

So calcium oxide looping is cheaper today but has less prospects of costa coins in the future.

Is that Did I understand that correctly?

Yes?

But you know, even so, if you're looking at kind of our graph that's in the note, if you guys have access to that, it could in the best case scenario, liquid absorption technology could still be competitive with some of the more expensive forms of electrochemical technologies by twenty fifty, whereas even though we're seeing solid sortment technologies start at a far higher first be kind cost realize a much more dramatic cost down by twenty thirty, and by twenty fifty you're still seeing solid technologies probably be on the more expensive side of all three.

Obviously there's margins of error of that.

It's such an interesting point because, I mean, fun fact about me is when I joined ANYF, and it was before we would be NF.

I joined any F as an intern in the summer of two thousand and nine, which really ages me, and I was looking at CCS, and so I started, you know, before I started my internship, googled CCS and learned what I could about it.

And one of the things someone tulted is, like, the great thing about CCS is it's built out of existing mature technologies, so it's definitely gonna work, which you know, some people might dispute.

But like the thing that now hearing you say this, this reflection is is that thing that was tiulted as a strength it's built out of mature, well understood technologies, is actually, in a way it's weakness.

I was about CCS.

It isn't about DAK but you know, obviously they're yeah, tod, but this is actually being built out of mature technologies.

Is only an advantage up to our point.

And if we're thinking about the long term prospects for something, maybe something exciting and new is exciting and new.

Also harder to project finance, then yes, yeah.

Of course, of course that's I mean, if it was all that easy, then we'd all be just getting rich off of whatever is exciting and you but let's talk about what's exciting and new electric chemic tell us about what that is and how it works.

And yeah, you mentioned it's so nascent.

Do we even have a cost idea for it?

Right now?

It's so nebulous, right, nobody really knows.

You've got to go to the label of the parking lot.

You have to go to the parking lot.

Ask people how much does it cost?

They're just trying to buy their groceries.

But your parking might be more expensive.

Most probably everything.

Nothing is more expensive than New York pokast.

Yeah, so I guess in our modeling, we don't actually start modeling for electro chemical until I think it's around twenty thirty one.

That's around when we're expecting to see the first of a kind of commercial.

Plant come online.

At that point, you know, we start the modeling and around seventeen hundred dollars per ton of CO two and then obviously is a function of demand, which it could be an entirely separate podcast, similar to what we've seen in the battery industry, we could see learnings around eighteen percent.

That's a function of modularity, that's a function of the system being extremely novel holorified batteries.

So I mean, it's a chemical process, right, and that eighteen percent learning is pretty standardized of learnings that we've tracked in other chemical processing, right.

I mean, I mean at BNF, we've looked at batteries for more than a decade and seen established these kind of learning rates for those kinds of technologies.

So that's why it's exciting, is it has that prospect of that exponential cost decline definitely.

And I think what's probably more exciting is the fact it requires seventy percent less energy, And especially when we're kind of I mean the narrative right now in the media and throughout I feel like most of the work we're doing at BINA is increased load from data centers, right, we don't have enough energy and scary energy security things like that.

And then you add DAK into the mix, where there has been a lot of negative pr recently saying these require too much, too much energy and the land use required if it's you know, to power deac, if it's solar wind, far too large and it doesn't even capture as many tons of CO two as it took to build the plant, right, which is you know, not necessarily true, unfair for a lot of reasons.

When when I asked you earlier about the cost of energy, and why I asked is it electricity or fuel is because everything you say is correct about you know, increasing load, and you know this might be something another competing thing.

But you know, my jam is I look at the power system, power markets, and whilst it's true that overall load is growing, at certain times of day there will be excess power.

And the ideal scenario is there something that can be useful that can be done with that power.

And the ideal candidate is something that is energy intensive but not too CAPEX heavy, so you don't need to be worrying about running it twenty four hours to sort of fully monetize your initial investment.

You can run it during the daytime when you know the solar is producing excess powercuse that's probably what it would be, it would be excess solar.

And so I was really asking because I'm like, is that a potential future for DAK or whatever can run on electricity and has maybe a lower proportion of its cost is in the capex rather than the energy costs.

Yeah, it's obviously for solid solid sorbit DAC and calcium oxa being impossible.

As you said, it's far too capitally intensive to even think about wanting to do that.

We did just have to run it all the time.

Yeah, yeah, one hundred percent.

We did run an analysis on that.

Which will be coming out in a second part of this cost now.

Oh nice, watch this space.

Yeah, we're teeing that up.

Also for electrochemical there are some companies that have said, hey, we're running on behind the meter power, or hey we've built in a sort of battery into the process of the system, and so we don't need power one hundred percent of the time, right, and we'll just use these reservoirs that we have of different chemicals sitting and that will be.

Able to kind of I guess power it.

So that has been floated.

I don't know if I can say, oh, yeah, we're definitely going to be able to see electric chemical running on intermittent renewables, because the economics of everything are just still so ambiguous.

For electric chemical in particular, that would be obviously the best case scenario.

Right, but it hasn't been yet been proven out.

Yeah, I think it's still pretty ify because again, right now, electric chemical it's still insanely gapply intensive, got it, you know?

And who's to say we don't see based on lower maybe demand signals.

Maybe there isn't this crazy growth of demand from all of these companies, you know, dack realistically or maybe more pragmatically, maybe we won't see that eighteen percent learning which we'll get it down to those kind of eighty dollars per ton levels.

Maybe we do still see electrochemical hovering at that two hundred dollars per ton range, which will mitigate that ability.

I think two hundred dollars a ton sounds expensive, but for something that is finessing the last little details of combatting climate change, that's not expensive.

You know if that's where it can get.

To, right, and then especially I mean we had the euets is at around one hundred dollars per time.

You don't need DAK to get to below one hundred dollars per time.

That's obviously, you know, if you're operating in a market with a compliance market, right, But in some instances you just kind of have to reach parody with that.

Thank you very much for this conversation, Brenna Casey.

I came at you with a bit of a barrage of unfriendly questions, and to be honest with you, if you hadn't given good answers, then the second half of this podcast would have been pointless, yeah, because we would mean, why are we talking about the cost of something that doesn't matter?

But I think you really did establish that it does matter.

And it's actually really quite an interesting topic, both from a technological point of view, but also you know, coming back to this idea success doesn't happen in straight lines.

How this nascent industry that let's say, let's be honest, it is facing some headwinds, but it also has friends in important places, it sounds like, so how does it navigate its way to the top, and I think all of this is so interesting and so thank you so much for coming on today.

And I guess as of next week, by the way, we're going to be sitting on neighboring desks.

So I suppose I should start calling you Brenner rather than Branna Casey.

But yes, thank you for joining today.

Thank you guys for hiring me.

Today's episode of Switched On was produced by Cam Gray with production assistance from Kamala Shelling.

Bloomberg NIF is a service provided by Bloomberg Finance LP and its affiliates.

This recording does not constitute, nor should it be construed as investment advice, investment recommendations, or a recommendation as to an investment or other strategy.

Bloomberg ANNIAF should not be considered as information sufficient upon which to base an investment decision.

Neither Bloomberg Finance LP nor any of its affiliates makes any representation or warranty as to the accuracy or completeness of the information contained in this recording, and any liability as a result of this recording is expressly disclaimed.