In this episode, Princeton PhD candidate Wilson Ricks, co-author of two papers on enhanced geothermal systems, discusses the potential for that technology to play a big role in the clean energy transition.
Text transcript:
David Roberts
In recent years, excitement has been growing about the potential of geothermal energy, which draws heat from the Earth’s crust to generate electricity. I wrote a couple of introductory pieces on it for Vox a few years ago (one, two) and they remain some of the most popular things I ever published there.
To date, geothermal has largely been viewed as an always-on (“baseload”) resource, like nuclear power, that needs to maximize its running time (“capacity factor”) in order to maximize its revenue. Basically, it has been seen as a clean substitute for coal and nuclear power plants.
However, new research from Princeton University's Zero Lab suggests that enhanced geothermal systems (EGS), which fracture underground rock to create their own reservoirs rather than relying on natural ones, can play a much more dynamic and valuable role — a role more like the one fast, flexible natural gas plants play today.
In a paper published in May, researchers show how EGS plants can store energy (for up to 100 hours or more!) and rapidly ramp their output up or down, allowing them to act as both storage and flexible “clean firm” generation in a decarbonized electricity system.
If an EGS plant can store large amounts of energy for long periods of time and crank up its output when it is most needed, its value exceeds previous estimates by as much as 60 percent. And if EGS plants are more valuable than previously recognized, a follow-up paper from the same team shows, it could end up playing a much larger role in a decarbonized energy system than previously envisioned.
I contacted one of the authors of both papers, PhD candidate Wilson Ricks, to discuss what the research found, what EGS plants can and can’t do, and the expanded role they might play in clean energy going forward.
With no further ado, Wilson Ricks, welcome to Volts. Thanks for coming.
Wilson Ricks
Thanks, David. It's great to be here.
David Roberts
You have done — or been involved in — some research here on some of my very favorite topics in the world, so I'm excited to talk our way through this. But just to begin with, to orient our audience a little bit, why don't you explain what the difference is between plain old geothermal power, which has been around for a long time, 70 or 80 years, I think at least, geothermal electricity, and what they're now calling "enhanced geothermal."
Wilson Ricks
Sure. And I think I'll probably start with just a description of conventional geothermal, because, to be honest, I think people don't know very much about it. I didn't know very much about it when I started this work, so I'm going to kind of take myself as a baseline for people who know something about energy. So geothermal for electricity generation, it's actually been around for quite a while. The first plant was built in 1904 in Italy, and that's at a place where they literally had steam just coming out of the ground. And someone had the great idea to like, well, we already burn coal to make steam, to generate electricity.
Why don't we just drill a well into this and hook a generator right up to it? So it's kind of progressed from there. We've had geothermal plants in the US since basically in the 60s kind of operating on that same principle. And all geothermal power is really quite simple. It's just taking naturally heated fluid, typically water or brine from within the earth's crust. And by crust I mean, like near the surface, and passing it through some kind of thermal power cycle. And there are a couple of ways to do that. If it's pure steam that's coming out, it can be just through a turbine, straight through.
There are other plants that you can build for lower temperature resources, maybe ones that are more what are called "liquid dominated," where what comes out is just superheated water. But all of them and all the conventional geothermal power in the world today relies on this very specific kind of geologic formation to work. They're all built over what are called "hydrothermal reservoirs," which are places where you have a confluence of three really important things. Those are heat, fluid, and permeability. The heat, of course, is the energy source. And so that's kind of a must. The fluid is usually in situ water and that's your energy carrier.
That's how you actually get the heat energy from where it is deep underground up to your turbine and use it to generate power. And the permeability is the kind of key thing here too, because the formation where you have this heat and fluid needs to have enough hydraulic permeability. That means basically the ease through at which water can flow around in it for the fluid that's there to efficiently extract heat and then be pumped up to the surface. And the confluence of these three conditions is really quite rare.
David Roberts
Is it plate tectonics going on down there?
Wilson Ricks
It can be a variety of things. It could be places where there are rifts. Like for example in Iceland, we literally have the plates splitting apart through the middle of the country and that's a big geothermal hotspot. Kenya has similar conditions. There can also be places where it's just kind of an upwelling of magma from deep underground creates a hot spot. That's what we have in Yellowstone and also at a place like The Geysers in Northern California, which is the largest currently operating geothermal complex in the world. There's over a gigawatt of capacity there.
David Roberts
Right, but the thing that ties all the conventional geothermal together is that these are naturally occurring phenomenon that you're just finding and then building a power plant on top of.
Wilson Ricks
Yes, they are naturally occurring, but as I kind of implied, they are pretty unique and they're pretty rare.
David Roberts
Right.
Wilson Ricks
In the United States, we have already tapped most of the high quality hydrothermal resource that we have. And currently, geothermal power generates a bit less than half a percent of our total electricity. The USGS. Does estimate that there could be up to 30 gigawatts of potential for conventional hydrothermal, geothermal power plants in the US. But most of that is going to be lower temperature than what we've already tapped to date. And that means that you get lower conversion efficiencies.
You have to drill more wells to get the same energy.
David Roberts
Right, this is something to remember throughout the conversation, is the hotter the stuff coming out of the ground, the more efficient the conversion to power.
Wilson Ricks
Yeah. And so, like I was talking about, The Geysers, that's a place where it's so hot that you drill into it and pure steam comes up, and it's something like 400 degrees Celsius. And so you get very efficient, cost effective power generation out of something like that. Whereas when you get down to the 200, 150 degrees Celsius, you have to drill more wells to get the same amount of energy out. And because of the way thermal power cycles work, the actual conversion of that heat energy to electricity is much lower efficiency. And so the costs kind of decrease with higher temperatures very quickly, and they increase with lower temperatures.
And so we're kind of getting to the back end of the high quality hydrothermal resource in the US. And so when you think about what's needed for economy wide decarbonization, where we have to decarbonize the entire electricity sector, we have to generate double the electricity we need today. There isn't really much more than a niche role for conventional geothermal because there just is not that much of it.
David Roberts
Right. Oh, I wanted to add one thing to your description. Maybe everybody already knows this, but it's worth just adding to complete the picture. The way these plants work is like sticking two straws in the ground. You're blowing into one straw, you're blowing right. Water down into these cracks and stores of hot water, and consequently forcing hot water and ore steam up through the other straw. Just so people have kind of a visual.
Wilson Ricks
Yeah, And so what you typically do is when you extract that hot fluid out of your production, well, you'll run it through whatever surface power plant you have.
David Roberts
Right.
Wilson Ricks
And then it'll be cooled. You'll have extracted the energy, and you'll just inject it back into the same reservoir. And that way, you don't actually reduce the total amount of fluid in the system, and it can go back to pick up more heat from the hot subsurface down there, and the cycle kind of continues.
David Roberts
So that's traditional. And it is, at least in the US, and in most places, largely tapped out. And there just isn't enough of it, really, to play a big role in decarbonization.
Wilson Ricks
Yes.
David Roberts
So then along comes the magic of technology.
Wilson Ricks
Yeah. Along comes, well, several concepts I don't want to say that EGS is the only concept out there, but I guess along comes the realization that even though the hydrothermal resources that we've been talking about, which are the confluence of heat, fluid, and permeability, even though those are very rare, heat is not rare. If you look at a map of temperature at depth of the United States, you'll see that there are large, large swaths of the country where if you drill down just a few kilometers, you get to very, very high temperatures at which you could economically generate electricity.
David Roberts
Right. So what's missing then is the fluid and the permeability.
Wilson Ricks
Yes. And all of these things are on a spectrum. So you could go from very, very high permeability to gradually, gradually lower. This ranges over orders of magnitude in the crust. But in general, you've got kind of a distinction between places where you could produce geothermal power just by literally drilling a well into it and pumping the fluid up in places where that's not really feasible. And so to actually make that work, to actually get all that heat, which is, by the way, just enormous resource, it's enough to power civilization for thousands of years. In theory, actually getting at that is really the key challenge.
And so the, I guess, idea behind enhanced geothermal systems is, well, we have the heat, we have the rock, we don't have the permeability or the fluid. And so we're going to add both of those. We're going to create that. And the way you do that is effectively by using very similar techniques to what have been developed and really successfully scaled in the oil and gas industry. You drill down and you create a series of fractures in this rock that would be normally so impermeable that you couldn't pass water through it.
David Roberts
Which is, let's pause to say exactly what they're doing with natural gas. It's exactly the thing that unlocked this enormous new resource of natural gas in the US is fracturing rock, creating that permeability so the natural gas could come up and out. This is just fracturing the rock, creating permeability.
Wilson Ricks
So water can flow through it.
David Roberts
Yes, so you can pump water through it.
Wilson Ricks
Yeah, And so what you would do is we have that two straws model from before, but now you basically you drill your first straw, you create that fracture network, you create pathways that water can flow, and then you try to make sure that those pathways intersect your second well. And so, at the end of the day, what you hope for is a system where you can pump water down one well, it will migrate through those fractures, and then as it goes, it will pick up heat, and then it will pass up through the second well and you can pump it back to the surface and generate electricity.
David Roberts
Right. So this is enhanced geothermal. It's basically creating a geothermal resource where there was none, taking an area where there's just heat, introducing fractures, pumping water down through it. The water picks up, heat comes back up through your other straw, and you're creating heat.
Wilson Ricks
Exactly.
David Roberts
And this is relatively new and relatively parasitic. I mean, I think it came out of or was juiced by, as you say, advances in fracking technology and other advances in oil and gas. So give us a sense, before we get into sort of the subject of that first research paper, just give us a sense of where enhanced geothermal is at. Are there lots of enhanced geothermal wells? Where do people view the potential of it? Where does it sit in the sort of technology curve?
Wilson Ricks
So it's quite early, though not in the conceptual phase, becpause it's actually been around. The idea of it has been around for quite a while. The first attempt to actually create a EGS — which is the acronym for it — system in the world, was begun actually in 1977 at a site called Fenton Hill in the United States, where they did attempt this. Of course, back in the day, this is decades and decades before the shale gas revolution. And so even though hydraulic fracturing did exist as a technology, it was very much not advanced to the state where it is today.
And so they had some trouble with it. There was issues actually getting the fractures to go in the right direction and connect the wells. They actually drilled the two wells first, and then when they tried the fracturing, it just didn't quite go right, and so the connection wasn't great.
David Roberts
How do you do the fracturing? I think people sort of envision, like dropping dynamite down a hole or something and blowing what creates the fractures.
Wilson Ricks
The way this works, and this is basically nearly identical to what happens in oil and gas, but there are some differences, is that you're basically taking a section of your well, typically isolating it so that it can hold pressure, and then you're just pumping as much water into it as you need to actually, literally get the rock to split from the pressure.
David Roberts
So it's just water pressure basically?
Wilson Ricks
Yeah, it's essentially just really high pressure water. And you inject enough water to get the fractures to kind of continue growing as long as you need them to. You also sometimes inject what's called proppant, which is essentially little, it can be something like sand or little particles. When that flows into the fracture with the water, it can actually prop the fracture open once the water pressure recedes. So your little sand particles actually provide kind of barrier to the fracture closing on itself again while still allowing enough pathways for things to flow through once you try to operate the system.
David Roberts
Got it. So we've just gotten better then, since then at sort of precision fracturing than fracturing where we want and in the direction we want.
Wilson Ricks
Yeah, we've gotten a lot better to the point where the productivity of these unconventional oil and gas wells is just like many, many times greater than it was before the mid 2000s. But these techniques have not really been applied to geothermal yet. The initial wave of exploratory tests of enhanced geothermal happened before this, and the advances that we've seen, paired with the renewed interest in all things clean energy, has led to kind of a second wind, where there are projects ongoing by the US. Department of energy, by private companies to try to leverage these advances and really make EGS work.
David Roberts
Talk about the earthshot.
Wilson Ricks
Yeah. So the DOE recently announced the newest addition in their, I guess, earthshot series, which are basically they build them as all of government efforts to bring down the cost of critical clean energy technologies. So they have one for hydrogen, they have one for solar power. And the goal of all of these things is really to set a very ambitious target for technology development and leverage resources by the DOE, by other branches of government to try to meet that. And so they announced one just this past month for EGS, where their stated goal is to reduce the levelized cost of electricity to $45 a megawatt hour, which is kind of a point at which it could reliably compete in future electricity systems.
David Roberts
Do we have a good enough sense of where it is now? I mean, do we have enough of them going to sort of have an average sense of how much that power is costing now?
Wilson Ricks
We do not have a great sense of that because there aren't really enough EGS projects operating in the world today, and none of them have been built in the decades since we've had all these recent innovations. But the estimated current cost is roughly ten times that amount, just based on so they build this as a 90% reduction in cost. That is ambitious from what is the question, because there isn't really a commercial EGS industry right now.
David Roberts
Right.
Wilson Ricks
But safe to say, this is very much an emerging technology and there are projects currently ongoing that are trying to bring it into the modern era. And I think once those start getting results, we'll really be able to see where it stands economically.
David Roberts
Yeah. Kind of bold to set a price target before you even know what the current price is.
Wilson Ricks
Yeah. And to be fair, there are ways to estimate the cost of these things because we know how much drilling costs, and we have a sense of how much it would cost to geothermal wells are different from oil and gas, but we know how much that costs, roughly. We know how much the surface plants cost because conventional and geothermal plants use them. Where the uncertainty really lies is how well you can actually engineer that artificial reservoir. And so getting that to be a high performance reservoir, where you get lots of fluid flowing through it, lots of heat being extracted, heat being extracted efficiently.
If you can get that right, then you're looking at some pretty competitive costs.
David Roberts
One other preparatory question. So when we think about geothermal, generally, the conventional way of viewing geothermal is as what they call a "baseload resource," which means it's kind of always running. So we think of nuclear power. I think most people at this point are familiar with the fact that once you build a nuclear power plant, you want to run it all the time. It costs a lot of money and is inconvenient to ramp it up and down. It's not very flexible. It's sort of an always-on "baseload" kind of thing. Traditionally, when people envision geothermal, they're like, well, the heat coming up from the rock is steady and so you have the sort of steady production.
You want to sort of maximize your capacity factor, maximize the time you're running to maximize your revenue. It's an always-on resource like nuclear. And before we get into why that might not be the case, let's just talk about why that might not fit well into the current electricity. So, why baseload is sort of not a good lane to be in these days.
Wilson Ricks
Yeah, this is something that people fight over. Who is the first to claim that baseload is dead.
David Roberts
That was in the 70s too, I'm pretty sure.
Wilson Ricks
Yeah, it's become a popular saying in energy circles. And the reason, I guess, to explain that I think we should first think about, like, what the kind of market for baseload power is in traditional power systems which is really that if you look at the overall demand for electricity. There is always going to be some. It will change when people turn on their AC during the day or get home and turn on their dishwashers at night. But there's always going to be a minimum. And that's literally called the baseload. And so it makes sense to have some generators that literally always run, like there is a market for that.
David Roberts
Right.
Wilson Ricks
The problem — maybe not a problem overall, but the problem for baseload — starts to enter when you have things like wind and solar into the mix. And that's because these are not dispatchable resources. They generate whenever the resource is there. So when you look at this from the perspective of your conventional dispatchable generators, things that say, and that includes the baseload ones, the market that they have left to fill at any given moment is effectively the original demand minus whatever's being generated by wind and solar.
David Roberts
Right.
Wilson Ricks
And that's what we call the net load.
David Roberts
Yes. And sometimes when the wind and solar is very vigorously going, that net load can go to zero.
Wilson Ricks
Yes.
David Roberts
Which never used to happen before.
Wilson Ricks
And that's really bad for baseload generators because their kind of economic proposition is that they have pretty high, upfront, fixed costs. So actually building your generator and having it there, keeping it up is fairly expensive. But because they run all the time and have a low running cost, they can effectively make that revenue to cover those costs over by just running for the entire year.
David Roberts
Right. They amortize it over time. And this is why just, again, to state the obvious, this is why natural gas has proven so popular and successful during this new era of power. Because unlike coal and unlike nuclear, natural gas combined cycle plants are actually quite good at ramping up and down to complement those swings in solar and wind power.
Wilson Ricks
Yeah, and so you've kind of got two issues at play which affect geothermal in different ways. One is the ramping, which is the fact that if you have wind and solar in the mix, those are changing and demand is also changing. And so if you combine those two things, you can have a very high rate of change in the net load. And that means you need things that can ramp quickly. Nuclear, coal, those aren't very great at that. Gas is. Geothermal actually is too. Modern geothermal power plants have very high ramp rates. A binary cycle plant can ramp up to, I think, 30% of its power per minute, which is extremely high.
The problem is that that whole rampability aspect is only part of the reason why baseload suffers when you've got wind and solar. The other half is the economics, which is just that they are not making money during the middle of the day.
David Roberts
Right. Which is why nuclear is why we're having all these fights about keeping nuclear plants open like they're not doing well on the market.
Wilson Ricks
Yeah. So with geothermal, the problem with geothermal in terms of deviating from a baseload operating strategy is that its fuel cost is zero. It doesn't actually save any money by not generating power.
David Roberts
Right?
Wilson Ricks
And so when you have times when the value of its power goes near zero because there's so much wind and solar in the system, it then just basically has to make up its costs by making more money during other hours, and it might not. And so it really suffers in that way. And all baseload generators have this issue.
David Roberts
Right. So insofar as geothermal is a species of baseload power, it is going to run into the same economic headwinds that nuclear and coal are running into for some of the same reasons. So that's all background. Which brings us to the research which suggests that geothermal can play a non-baseload role, can in fact, do other things. So tell us what those other things are and walk through it for a non-engineering audience, how it can do those things.
Wilson Ricks
Sure. The best way to start is with a little history lesson of that original EGS test project that I talked about in the US called Fenton Hill. They actually did something really interesting. It was one of their very final experiments towards the end of that project, they did what they called a load-following test. And what that test entailed was they had their system running in kind of its steady state, where they were injecting water into the injection well under pressure. It was being forced by that pressure through the fracture network that they had created. And they did successfully create a connection, even though it wasn't a very good one, but they were able to do some tests on it.
It was functional to that extent. So the water was migrating through these fractures and back up the production well. And they were kind of injecting in a steady state, producing at a steady state. And then what they did was they partially shut in the production well. This is by basically increasing the back pressure at the top of it to reduce the flow rate. But they maintained the flow rate at the injection well.
David Roberts
Right.
Wilson Ricks
So what they were effectively doing was pumping more water into that subsurface reservoir than they were actually extracting. And what they noticed was that the pressure in the reservoir rose. Which makes sense, because the one thing that is interesting that distinguishes an EGS reservoir from a conventional reservoir is that the EGS one is at least theoretically surrounded by impermeable rock, because you created it in impermeable rock.
David Roberts
It's bounded.
Wilson Ricks
Yeah. And so when you're pumping water into it, the only way out for that water is out the production well.
David Roberts
Right.
Wilson Ricks
And so if you're not taking it out the production well, it's not going anywhere. You'll have an increase in pressure that comes with slight compression of the water and the rock in that reservoir, but that water is effectively being stored down there under pressure. And then they had this going for a period of a few hours, and then at the end of that period, they opened up the production well. They allowed way more water to flow. And what happened was all that accumulated pressure drove all that fluid that had also been accumulated up the production well and effectively allowed them to produce fluid at a much greater rate than the steady state for a period of time.
David Roberts
Right. So it's like you're partially sticking your finger over the second straw to build up pressure, and then when you take your finger off, you get this period of higher than average production.
Wilson Ricks
Yeah. And so at Fenton Hill, they did this repeatedly over a period of like multiple daily cycles. They found that they could maintain a much higher production flow rate for periods of many hours during those high flow periods. And what this is effectively doing is because you're producing less during some periods and producing more during others, you're effectively shifting when you're producing that same energy. And that's basically energy storage. You're not necessarily taking in power from the grid and storing it and discharging it later, but you are effectively taking power that you would have produced, storing it, and then discharging it later.
So effectively the same thing through a different means.
David Roberts
Right. So one question about this, though, is that my understanding of enhanced geothermal versus geothermal is that there's not really a clean line between them. So what happens is as you move farther and farther away from natural fractures, you get sort of less permeability, and then you can, with enhanced geothermal, kind of enhance that permeability. So, in other words, have we actually gotten to the point in enhanced geothermal where we're doing a purely new set of fractures in otherwise unfractured rock? Or are we still kind of in that period where we're on the fringe of traditional geothermal fields with just sort of not no permeability, but less permeability?
Wilson Ricks
So most of the projects undergoing right now are getting into the point where, for example, the Forage Project by the Department of Energy, that is in solid granite, and that's a rock where there is no water that's going to be flowing in that naturally. There is a gradient, like you said, and there are projects that kind of bound the edges of that. There was one in Canada called Deep, which I think they were planning to do fracturing at one point, but they found that the natural permeability was actually higher than they expected slightly, and so they end up just going entirely without it. And so there are projects that are kind of on that edge.
David Roberts
And I'm just assuming that this technique will work best when you have a completely bounded set of fractures versus if you're still working with some natural fracturing, you have some leakage.
Wilson Ricks
Yeah. So that's a bit of an interesting question because actually, right now we aren't quite sure what happens at the boundary. We did some sensitivity tests in this paper, which we'll get to talking about basically exploring the impact of different variations in subsurface parameters on the actual performance of this. And we found that when the innate permeability of just the surrounding rock was higher, this actually performed better in a lot of cases. Now, what happens at the limit when you get into, you know, such high permeability that the pressure effectively just essentially evaporates away? That's what we could see at something like the Geysers, where they actually have had some experience with this, the Geysers being a fully high permeability, steam dominated conventional geothermal field.
They have some experience actually shutting in the wells there, and they do see a puff in additional production at the end of those shut ins. It just happens that the amount of energy they recover is something like 15% of the amount that they would have produced if they hadn't shut it in. Right.
David Roberts
Because it's leaking away. The pressure is leaking away in that high permeability.
Wilson Ricks
Yeah. And so there seems to be a kind of boundary point at which the performance stops increasing as it gets a higher permeability. The reason it does is because what we observed is that. In very, very low permeability, you cannot actually inject that much fluid into the reservoir before the pressure starts to really rise because there isn't actually that much space in the fracture network. As you increase the permeability slightly from that, you get an effect where water actually is able to leak off from the fractures into the surrounding rock a little bit as you increase the pressure.
And so the effective storage capacity of the system actually increases a little bit. And so figuring out where that effect ends and where the leak off effect becomes dominant is something that we definitely like to look into, but have not gotten to at this point.
David Roberts
What we have now then is an enhanced geothermal plant that can a. store energy, right? Because you can take energy from the grid to run that pump, to pump that water down and increase that pressure. So you're effectively pumping energy down there and storing it until you uncap that other straw. So you're storing energy and you're also ramping the output of your plant up and down. So you are basically no longer baseloading there. So explain the kind of the significance of this to the value of that particular plant.
Wilson Ricks
So this is geothermal basically. Well, it's almost acting as a hybrid. It can act as a baseload resource. Like you have the option to just run all the time. But what's really interesting is you can effectively also be an energy storage device and a load-following plant and it's kind of all in the same system.
David Roberts
Like a natural gas plant with a battery attached.
Wilson Ricks
Yeah, though I will say I want to distinguish this from traditional grid connected energy storage because again, aside from the pumping power which you just described, it's not actually absorbing electricity and discharging it later, it's effectively just changing when it produces its power, which from the grid's perspective looks almost the same. If you have a baseload resource that turns off that looks the exact same as if you had a baseload resource and a storage device on top of it that started absorbing all its energy, right, which is effectively what we have. And the reason this is important is that if you have a geothermal plant that could do this, it can reduce its output potentially even to zero, depending on certain physical constraints during the middle of the day when the price of electricity is near zero in some future with lots and lots of solar power everywhere. So it effectively isn't wasting its energy during that time and then instead it discharges its energy, say during the evening ramp, which is when, you know, solar power starts to go away, people go home, they turn on their, you know, TVs and their washing machines and all that stuff.
And so you need a. you know, really fast ramping and b. a lot more power from these dispatchable resources. And now, because like I said before, these modern geothermal plants tend to be quite flexible by themselves. It can ramp quickly and it can take all the power that it would have generated during the middle of the day and dump it all on the grid, grid at night. And that allows you to really increase the value that you get out of that same amount of fluid that you're producing.
David Roberts
And in the first paper, just to put a number on it, you say it increases the value of that geothermal plant by as much as 60%, which is not that's not a marginal effect.
Wilson Ricks
Yeah. So this can be really important in places like, for example, Southern California, which is where we got the price series data for that particular run. That's a place where you've got tons and tons of solar on the grid. You even have long periods of negative electricity prices. Actually, you get paid to effectively you stop losing money by being able to ramp down. Those kind of periods of very low prices are punctuated by periods of very, very high prices where the sun goes down and you need something else.
David Roberts
Really rapid ramps happening in California.
Wilson Ricks
And so in systems like that, which is probably what a lot of future electricity systems are going to look like, a strategy that shifts your power to the high price periods rather than producing all the time is way more valuable than one that's just standard baseload. So if geothermal plants could actually operate like this, they would make significantly more money, which means that they could either see much greater deployments or get away with having higher upfront costs, potentially, while still making economic sense.
David Roberts
I remember now people used to, or I guess still occasionally talk about sticking giant battery banks on nuclear plants, basically to do the same thing, to make the same thing. You're not going to get more net power out of the nuclear plant, but the battery allows you to shift when you get it.
Wilson Ricks
Yeah, that's actually the reason behind a lot of the original construction of pumped hydro storage in the US. It was meant to effectively deal with the same issue with large nuclear plants and shift their power to other periods. Now, I think the economic need for that kind of thing will only increase because back in the day that was only based on just the changes in demand. Now we're dealing with changing demand and changing wind and solar output. Yeah.
David Roberts
So the value of flexibility is going to go way up. And another aspect of this that you point out in the paper, which is sort of logically obvious, but worth underlining, is unlike in the nuclear case where you have to pay a very high premium if you want to add a giant battery bank. When you have your enhanced geothermal plant built, the battery is effectively already there. It's zero additional capital to store energy. Right?
Yeah.
Wilson Ricks
So that's the really interesting thing about this potential mode of operation. Is that the reservoirs that we simulated during this work were not specially designed or enhanced in any way to make this storage possible. There was no deviation from what you would normally do to just create an EGS power plant in the subsurface, at least. And so you're literally just taking advantage of the natural properties of a confined geothermal reservoir. And you don't have to build a large additional hydro reservoir or something to make this storage possible. What you do have to do is make some adjustments to your surface facilities.
So, for example, if you're suddenly producing at much higher flow rates during certain hours of the day, you can't take advantage of that unless you've built a bigger generator at the surface.
David Roberts
Right.
Wilson Ricks
And so that is a cost that can kind of limit it, because geothermal power plants tend to be actually fairly expensive. But that's really your constraint. And the important part is that that's a constraint on your power output, but it's not a constraint on your total energy storage.
David Roberts
Right.
Wilson Ricks
Which is really what you get out of your reservoir. And that's the thing that is particularly valuable when you start getting more and more variable renewables in the system, is that long duration storage, which we find that these reservoirs may be quite adept at providing.
David Roberts
Yes, this is something Volts listeners will be very familiar with, the need and the sort of challenges of long-term storage. And here, there's effectively no limit on if you can create your own reservoirs, right. If you're creating your own fractures, there's a lot of solid rock down there, so there's no obvious limit to how big of a reservoir you could create. And the bigger of a reservoir you create, the more storage you get and the longer you can hold it.
Wilson Ricks
Yeah, and so this isn't even something we've looked into yet. But there may be opportunities to effectively optimize the actual size of your EGS reservoir to provide optimal storage capacity, in addition to just the thermal performance that you would need to actually generate electricity in the first place.
David Roberts
One question I know a lot of people have is what is the sort of round trip efficiency here? If I shove a bunch of energy down and I'm holding it effectively underground under pressure for a period of time before allowing it back up through my other straw, how much am I losing over time?
Wilson Ricks
This is something that's particularly interesting, because what we found is the efficiency. Well, for one, it can be variable. How this was done was that we effectively took initial kind of simulations of the reservoir operating and then essentially turned that into an optimization model that would allow us to effectively keep track of those behaviors, but still optimize the plant operations as a power plant. And we saw that it would actually potentially take efficiency hits intentionally to increase its revenue if that was the case, if it could gain an advantage by doing so. But what we found was that the greatest sources of loss are likely to be the need to, I guess, keep the reservoir maintained at a higher pressure during periods when you're charging it up.
So if the reservoir pressure is higher, that means the pressure you need to put on your injection pumps to pump water into it is also higher, which means that the power that those pumps take is also higher. And so rather than kind of an efficiency loss on charging, there is this kind of increase in power that's just kind of constant as long as you have the reservoir charged up. So it's interesting kind of compared to traditional storage devices that might have just losses on charge and discharge, but it seems like the efficiency could be anywhere between high 60% and even up into the 90% round trip efficiency range. And this is from the perspective of a long duration storage device that's really quite good because if you look at something like round trip hydrogen storage efficiency, that's in the sub 40% range.
And the figures that we were able to come by for EGS were competitive with lithium-ion batteries or pumped hydro, which are kind of the cream of the crop in terms of grid storage technologies today.
David Roberts
Right. And just as we're talking, I mean, I think everybody who's listening will get the same sense, which is that this is so new and so unexplored that opportunities to improve these numbers are just going to be all over the place as you build more of these plants, like just improve the efficiency of the whole process. Since it's almost all theoretical. It's almost all in our heads. There are nothing but opportunities to improve performance on the margins. Yeah.
Wilson Ricks
And so I will caveat that and say that this is also because this is all theoretical, because we haven't actually built one of these yet. I do not want to say for certain that this is the key to all our problems that's going to make clean energy happen by whatever date.
David Roberts
Would you say it's a game changer Wilson?
Wilson Ricks
So if it works potentially, that's really what I think is the kind of major result from this work, is basically showing that if this works, as the stimulations imply, it could be a really big deal for EGS. And of course, EGS itself has to work in the first place.
David Roberts
Yes. So let's talk about that then. So if EGS works, and then we get this storage, this sort of delayed production storage thing to work, this is what your second paper gets into, which is assuming this works, and then you have these EGS plants serving as effectively battery flexible plant hybrids, like, really? Like a Swiss army knife. Like everything you want. Everything you want in a plant. If you can get these EGS plants working this way and then you model the sort of evolution of the power system in light of those working at the costs you're estimating, what will this mean for a. for the expansion of EGS, and then b. what will it mean for the larger project of decarbonization?
Wilson Ricks
Yeah, so that was the subject of the most recent work, where we effectively, once we kind of established a baseline for whether this added value at all for a single enhanced geothermal plant, we really wanted to explore what its potential role could be and what its impact could be on the long term value of EGS. Again, assuming EGS works, this is kind of looking at what is the prize if we can get the technology development down, and looking at how it impacts the long term deployment of geothermal power, and just overall how the decarbonization of electricity in the United States could be changed by the addition of this technology to the mix. And the results were really quite significant at kind of a baseline drilling cost, assuming that the actual reservoir development. So really EGS itself is successful.
We found that baseload EGS might be deployed at somewhere around ten to 15 gigawatts in the western United States in a completely decarbonized grid.
David Roberts
Give people some sense of that scale, how does that stack up to other sources?
Wilson Ricks
Yeah, so it was less than 3% of maximum demand in the system, and so it runs all the time. So it does generate more power than a lot of other sources. But overall, it's really still quite niche at that point. By comparison, the exact same technology costs, but allowing it to operate flexibly, like we talked about here, we get up to something in the 50 to 70 kilowatt range, which at that point is really playing a significant role. And this is still at fairly high cost. This is at capital costs over $5,000 a kilowatt, which is typically not super competitive for a baseload generator in these kinds of systems you're competing with.
This is a levelized cost in the $50 to $60 to $70 range compared to wind and solar that are at like sub-$20. So the fact that you could get deployment that high is really significant. And it's really because even though the costs are high, the flexibility raises the value by so much that it still becomes a good economic proposition. And that's really the other side of the equation.
David Roberts
And the value of flexibility itself rises and rises and rises as the system gets cleaner and cleaner.
Wilson Ricks
Yeah, and so that's why we think that this may be important to look at decarbonized electricity systems, because in those cases, if you're relying super heavily on wind, solar power, these kind of changing net load that we've talked about, that just becomes more and more and more extreme.
David Roberts
Right.
Wilson Ricks
We also found that in addition to just increasing the amount of geothermal deployed, just adding the ability to operate flexibly with all else equal has a pretty large effect on just the total cost of the electricity system, something like up to ten percentage points, literally just from that one tweak. And the reason for that is that if you have a geothermal plant that no longer has to operate as baseload that can effectively shift its energy to times when it's most needed, then you effectively don't need to have other resources that might have been filling those roles.
David Roberts
Normally, yes. Like natural gas with CCS.
Wilson Ricks
Exactly. So rather than building geothermal, like a little bit of geothermal to meet your baseload, whatever kind of is left of it, and then adding some natural gas plants with CCS on top of that to meet the load-following demand. And then maybe some peaker plants on top of those and some long duration storage. If Geothermal can do those things too, then you need less of those other resources or potentially even none of them.
David Roberts
Right?
Wilson Ricks
And because you've got one thing now delivering that range of services, the overall cost of the system ends up being much lower. Because when you really look at the breakdown, those kind of what I call balancing resources that basically make up the rest compared to wind, solar, they account for a disproportionate amount of the total cost of the system. And so anything you can do to lower that cost has a big impact.
David Roberts
Right, right. So, yeah, I mean, people, you know, I think a lot of fans out there of the 100% renewable system and there's been a lot of talk about how to get that. But I think people maybe don't appreciate if you start with a 100% renewable system and add just a tiny increment of clean firm right, balancing resources that you can crank up and down on command, you start reducing the overall cost by huge increments, just with a little bit of flexibility. So having some cheap and clean and carbon free clean firm resources is just huge.
It is the single most valuable piece, I think, of a decarbonized electricity system. And that's now what we're talking about, the role we're now talking about geothermal playing, which is just amazing. So to start wrapping up the sort of modeling you did of the effect it could have on the overall price of a decarbonized system and on the deployment of EGS, as you say, just sort of assumes existing drilling costs. But I think we can assume that if EGS caught on and people started building a bunch of plants, these plants are not nuclear plant scale projects, right?
They can be reasonably modular, so you can build a lot of them, so you can get that learning effect going. So presumably we would see rapid falling costs if we started building a bunch of these. So I just wondered, in your modeling, is there a sort of a utopian scenario where this gets super cheap? What's sort of outer bound of what this could do?
Wilson Ricks
So we did look at one of those. We had kind of an advanced drilling, lower bounding scenario on the cost of EGS power. Assuming that this is one of the real potential promises of EGS as a technology, even compared to conventional geothermal, is that it is potentially highly scalable and replicable and you can really potentially leverage those same learning effects that we've seen in a lot of other technologies.
David Roberts
You're not depending on natural features of the rock anymore. You can operate anywhere with a solid rock. So you can standardize basically.
Wilson Ricks
Yeah. So rather than building your little boutique ten megawatt plant because you found a little boutique ten Megawatt hydrothermal reservoir, you can instead build these things, say, 100 Megawatts at time, get economies of scale on the individual projects level, and then also do that exact same 100 megawatt project a bunch more times in that same geologic basin that you're working in. And because you've already demonstrated that whatever technique works best in that rock formation in the first few wells, you can then in theory, get it much more cheap going on from there.
David Roberts
Right, and as a side note, just the land demands of this is much lower than for solar and wind too. Like the footprint of these plants is not obnoxiously large.
Wilson Ricks
And that's because it's really all in the ground. There's a pretty significant underground footprint. No one really cares about that.
David Roberts
No underground NIMBYs.
Wilson Ricks
Yeah, but we did explore kind of the taken to its logical conclusion, like how cheap could this get and what would happen? And it really has an extremely large effect on the cost of decarbonized electricity, even if we have cases where we assume very optimistic technology development cases for other potential clean, firm technologies. But even then, low cost EGS that can also operate flexibly is really kind of I don't want to say the silver bullet, but ...
David Roberts
It's the golden ticket.
Wilson Ricks
It really does kind of do it all. It effectively is a kind of catch all complement to wind and solar power, where they provide really cheap electricity when they're around and then your kind of cheap, flexible EGS makes up the rest.
Yes.
David Roberts
This is really the piece of the puzzle that has been lacking, right? I mean, this is exactly the right shape to fit in where there was a piece missing.
Wilson Ricks
Yes. Now I said silver bullet. It's not really the silver bullet on a like kind of economy widescale or at least a continent widescale. Because even though EGS is at those costs, it would be quite cost effective. There are still places where it's better than others in the Western US, which is what we looked at, there's really high grade heat resources fairly close to the surface. This starts getting more difficult when you get into the eastern US, where the crust is much older. There are still hot spots, places like West Virginia where this might be potentially scalable, but it's unlikely that barring much, much deeper drilling at much, much lower costs, EGS is going to be something where we can literally deploy it absolutely everywhere.
I would love for that to be the case. And the farther down you go, you will find heat. So there is a pathway there.
David Roberts
Well, this brings up my sort of I've got to stop calling things final questions because I always render myself a liar. But I wrote a couple of big articles on geothermal last year in Vox and talked a little bit about the sort of technological frontier. One of the technological frontiers in geothermal is going deeper, as you say, trying to get down to super deep. Because, as you say, a. once you go super deep, there's heat literally everywhere. Like if you can get super deep, then you really are unlocking an anywhere resource. And also if you get super, superdeep to where temperatures are super-critical, water is at its super-critical stage, which in terms of physics, I have no idea how to describe, but in terms of performance, you get more than incremental increases in the amount of power you can get out.
You start getting sort of exponential increases in the level of power available. So superdeep is really intriguing. And the other sort of technological frontier is what's called closed-loop geothermal where instead of just where your straw goes down, is just a continuous straw all the way through. So there's no water going into the rock. The water remains contained in the pipe and you pump it down in the pipe and it heats up in the pipe and you pump it back up in the pipe and get the heat out of it. This is Eaver, the Canadian company is doing this.
Does this apply, this change in operations that would allow a geothermal plant to become storage and flexible generation? Does that apply to these super technologically advanced versions?
Wilson Ricks
So this is a really interesting question, especially when you start to get into the extremely high heat scenarios, because we're simulating water in its kind of condensed subcritical fluid state where it behaves normally. And so what happens when you actually start to get to the extremely high temperatures and pressures where you'd have supercritical resources is not something that we've explored at this point. And it's a really interesting open question, because that is really the if we want to think of Geothermal as having a fusion power-esque end game, if we can make these initial resources work, say we start like you. Said at the places where it's kind of near a conventional site and it's easier to make the fracturing work.
And then we move out to these deeper, hotter resources, if that can somehow keep going to the point where we're getting to many kilometers deep, hundreds and hundreds of degrees Celsius. Then you get into this point where it potentially, if you can make things work at those temperatures, becomes an extremely cost effective resource in a lot of places in the world. Now, how it can be then run as a flexible resource at that point is unclear. And it may very well be possible, but it's almost besides the point because it would be so cheap that you might not even need it.
That's really why I compare it to fusion power because it's kind of the promise of abundant clean energy anywhere. The farther we get down, there's so much energy in the Earth's crust that it is really unlimited. If it can actually be ...
The promised land. We should pause and emphasize that it's super hard to do stuff melts.
Melts very easily.
David Roberts
They're using lasers and sonic blasts and microwaves now. This is like beyond, beyond the frontier stuff.
Wilson Ricks
Yeah. So we're trying to make these oil field hydraulic fracturing, horizontal drilling equipment work at 200 degrees Celsius. Getting up to 500-600 degrees is kind of an entirely new ballpark and we don't actually have any of the stuff we need to do that yet.
David Roberts
Yeah. Bespoke materials and bespoke equipment.
Wilson Ricks
Yeah, but the prize is potentially very large. Going back to the closed-loop systems, that's a really interesting concept that relies on kind of a different economic approach from EGS. EGS is basically based on the premise that you want to have as much surface area contact between your fluid and the rock you're extracting heat from as possible. You want to basically get the most bang for your buck in terms of number of wells drilled and number of amount of heat you can extract. Closed-loop kind of takes an alternative approach where it says essentially, well, the fracturing for EGS is going to be really hard and it's unclear the EGS itself is ever going to be technically feasible.
And so rather than dealing with that, we're just going to not fracture at all. We're just going to pass a bunch of pipes through the subsurface, basically a giant heat exchanger. And rather than having to deal with water flowing out of our wells into the fractures into the other well, we're just going to have a steady flow and generate electricity that way.
David Roberts
It really is like if people are familiar with ground source heat pumps in their backyard, that's basically the same way they work. They just run water through pipes underground. This is just way deeper, way hotter, but more or less a similar idea.
Wilson Ricks
Yeah. And so the downside of that approach is that pipes are significantly less efficient at extracting heat than fractures just because the surface area to volume of water flowing through them is much, much lower. Fractures are very, very thin and very wide and so you have a lot of exposure to the rock and so the thermal extraction is less efficient. And so to make closed-loop work, you need a lot of pipes. So it's kind of pathway to economic viability is that the cost of drilling needs to be really low, which could be achieved through some of these advanced drilling methods that you hinted at with millimeter wave energy drilling and that kind of stuff.
And you can't do the same thing with closed-loop that I've been discussing with EGS, where you deal with the alternative pressurization and the in reservoir energy storage, but you can actually operate those flexibly too. There's a concept whereby those systems effectively are limited by the total amount of heat they can pick up, so they're not actually extracting all the heat from the subsurface as the water flows through. And if you were to slow down your flow, it would actually pick up more heat as it traveled through the pipes. And so usually they'd find kind of an optimum point where the amount of heat you extract and the amount of water you actually get out is best.
But during the middle of the day, say, if the price of electricity is very low because there's lots of solar power, they could effectively stop flowing the fluid in their wells and it could pick up a lot of heat. And then when you evacuate that fluid that had been sitting in those wells, the higher temperatures will allow you to generate more electricity. So there are actually concepts, and this is kind of telling, that people investigating all these different geothermal concepts are also thinking very hard about ways that they can be flexible and ways that they can follow-load, because that is such an important capability for a resource that would otherwise be constrained to being baseload.
David Roberts
Yes, again, referring back to the first paper, it's a potential 60% increase in value, which is huge. And that's not even including potential technological improvements, which, as we say, will ensue if this catches on. Okay, so this is very exciting. Geothermal has, at least in theory, the opportunity to become really like a Swiss Army Knife here. Something that can hold energy, large amounts, potentially large amounts of energy for potentially long periods of time, which is huge, and rapidly ramping, which is also huge. So just like the perfect puzzle piece here. But as we say, this is largely theoretical at this point.
So really to finish, tell us sort of like you're out there working with is it Fervo to show this working? Right, so where are we going to see this happen first?
Wilson Ricks
Yeah, so this is the really cool thing, is that based on the kind of initial modeling work that we were able to do, and this was in collaboration with Fervo kind of from the outset.
David Roberts
Fervo is an enhanced geothermal company, just so everybody knows what we're talking about.
Wilson Ricks
Yeah, and so they helped us out with the actual reservoir simulation because they, of course, have expertise in that and we're the energy modeling people, but we were able to kind of demonstrate the potential value of this and then roll that up into a proposal for the ARPA-E Open Solicitation in 2021. And that's basically when ARPA-E comes out and kind. Of essentially offers up money to anyone with any kind of crazy energy idea. Anyone can write a proposal in just about anything and we were able to get a grant to actually try to demonstrate this. And to be clear, this isn't entirely untested in the real world.
Those original load following tests at the Fenton Hill project were essentially the inspiration for this project because that was a really interesting capability that they did demonstrate. And then it was kind of forgotten about for 20 years because EGS development kind of fell by the wayside and there also just was not as much of an economic advantage to load-following back then. And that's all changed now. But we're now trying to actually do a field demonstration of this at a real site.
David Roberts
A real existing EGS plant, or are they building one to do.
Wilson Ricks
This a like kind of pilot project that they have already in the works. So this is kind of in parallel to it's also something that honestly, if we wanted to go to like the Utah Forge EGS test site that the DOE is putting on, they could do similar tests if they were interested. So the goal is to actually have a real world working system and see if this works and see what kind of constraints there may be on it, whether the behavior is predictable. We want to make sure that if we continually pressurize and depressurize the fractures, that doesn't cause things to change down there, make this behave in ways we don't expect.
David Roberts
Yeah, like people have noted, people from the oil and gas industry have noted that oil and gas operations work quite hard to not have large variations in pressure because large variations in pressure screw things up. So it's maybe not as simple as we've just sort of described it here. There's a lot of factors involved.
Wilson Ricks
Yeah, that's another thing. Like I put in place limits on the pressure that the model was able to reach when I was actually doing the modeling work on this to effectively acknowledge the fact that there may be you know, you don't want to ridiculously over-pressurize this system cause, you know, seismicity or something like that. And you also don't want to have it too de-pressurized because then your fractures might close. And so there are going to be limits on the real world operations based on some of those phenomenon. And the goal of these projects is really going to be to see what those are, recalibrate the models, figure out what kind of value we're actually looking at in both the near and long term and see if this can be a thing.
David Roberts
Yeah, those are exactly the kind of problems that repetition will solve. Right? I mean, this is what the learning effect is for is precisely sort of like what's the optimum arrangement so that you can vary pressure as much as possible without breaking things, et cetera, et cetera. All these things will be experimented on and tested if you kind of get the ball rolling.
Wilson Ricks
Yes.
David Roberts
Well, awesome. Wilson, so cool. I love this. I love this research. I love this idea. I mean, if you're like an electricity system decarbonization geek, this is like all of us have been trained at this point, not to say silver bullet, not to say game changer anymore, and for good reasons. But this really has a ton of potential. So thank you for coming on and explaining it.
Wilson Ricks
Thank you very much for having me. It's been great.
David Roberts
Thank you for listening to the Volts podcast. It is ad-free, powered entirely by listeners like you. If you value conversations like this, please consider becoming a paid Volts subscriber at volts.wtf. Yes, that's volts.wtf so that I can continue doing this work. Thank you so much, and I'll see you next time.
Share this post