Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
1 | Casey Handmer — Mining the Air
705 Plays1 year ago
Casey Handmer is the founder of Terraform Industries (TI).
TI is pioneering air-to-fuel technology to manufacture methane (natural gas) from the air. Currently, we continue to extract enormous quantities of hydrocarbons from the crust, burn them, and release carbon dioxide. Instead, TI wants to mine the air: displacing the transport of carbon from the crust to the atmosphere.
Casey Handler has a PhD in theoretical astrophysics from Caltech, he’s worked at Nasa’s JPL and on Hyperloop One.
His blog is at caseyhandmer.wordpress.com
Twitter at @CJHandmer
Terraform Industries website: terraformindustries.com
For a transcription and further references see multiverses.xyz
My thanks to Mark Shilton, Sam Westwood and Maciej Pomykala for help with this episode.
Recommended
Transcript
The Carbon Dilemma and Hydrocarbons
00:00:00
Speaker
There is too much carbon in the air. Yet we continue to take carbon out of the ground in the form of hydrocarbons, burn them and put them into the air. And we do that because hydrocarbons are really useful fuels. We use them in our cars, in our planes, in the production of concrete and of the smelting of metals. So it's not easy to sort of solve our addiction to hydrocarbons.
Terraform Industries' Vision with Dr. Casey Hanmer
00:00:25
Speaker
But what if, instead of extracting them from the Earth's crust, we could extract them from the air itself? This is the plan of Dr. Casey Hanmer. He's the CEO and founder of Terraform Industries, who intended you just that, to mine the air.
00:00:41
Speaker
Casey has a PhD in astrophysics. He's worked at NASA's Jet Propulsion Laboratory. So he's literally a rocket scientist. And in fact, many people will know Casey for his wonderful blog on space, on the industries around space, SpaceX and developments there. He also flies planes. So he's a really, really accomplished guy. And I find it
00:01:09
Speaker
great cause for optimism that someone like Casey is working on a problem like this.
Economics of Air-Derived Hydrocarbons
00:01:14
Speaker
We're going to talk about how a combination of a centuries-old technology, the Sabatier process, in conjunction with electrolysis, in conjunction with the game-changing economics of solar power, how costs continue to come down, and as we deploy more solar power, we expect them to come down even further. How these things are going to combine and make it cheaper to extract our hydrocarbons from the air than from the ground.
00:01:43
Speaker
I'm James Robinson, and I live on a planet where we have a carbon dioxide concentration of about 420 parts per million, and our atmosphere does too high for us. But in another world, it's getting better. This is Multiverses.
00:02:15
Speaker
Casey Hammer, welcome to Multiverses. Thank you so much. It's great to be here. Thank you. So I've worked in startups for about a bit over a decade, and I've heard a lot of talk about companies who want to change the world, which typically means that they're going to put some pixels on a screen and sort of tweak human behavior. Your company, Terraform Industries, as the name suggests,
00:02:45
Speaker
is trying to change the surface of the earth. So what do you do? Well, if anything, we're trying to keep the surface of the earth the same, that is at a fairly steady temperature. But I guess in a single sentence, our overarching goal is to smoothly displace fossil oil and gas as a primary energy source for humanity by providing a better and cleaner alternative that is
00:03:16
Speaker
cross-compatible and cheaper. And so that alternative is methane, I guess, produced by renewable energies, specifically solar. And my Shiki, Change the Surface of the Earth, is kind of envisioning a world that's covered in solar panels, which I think is sort of a part of the plan. So yeah, could you unpack it a little bit more?
00:03:44
Speaker
How do you create this alternative fuel?
Synthetic Hydrocarbon Production Process
00:03:49
Speaker
Yeah, that's exactly right. And to be clear, at world scale, the price of getting off fossil fuels forever is deploying a lot of solar panels. But ultimately, very few on land that is already utilized by humans, and somewhat less maybe 10% of the total land that humans already use for agriculture. So the net impact is actually
00:04:11
Speaker
Rather low possibly even negative in terms of you know, the nuts and bolts chemically speaking
00:04:17
Speaker
humans use hydrocarbon fuel, which is gasoline or petrol for their cars, and diesel and kerosene for their trucks and planes, and natural gas, often for heating, electricity, cooking, and other purposes. Natural gas is also used in some vehicles. Natural gas chemically is methane, or methane, or CH4. It is the shortest, the simplest of the hydrocarbons. And that is the reason we started there. But our process in principle can extend to make
00:04:45
Speaker
any hydrocarbon or even a plastic materials that humans depend on and use in such vast quantities to safeguard the health and wellbeing. And so our principle is to produce those synthetically. So in order to make a hydrocarbon synthetically, you need both a source of hydrogen, which is the hydro part of hydrocarbon. And for that, we simply split the hydrogen off water, H2O, using a process called electrolysis. This process has been known about since the 1800s.
00:05:14
Speaker
and commercialized at scale since the early 1900s. And then we also need a source of carbon. And historically speaking, people have used coal or trees as a source of carbon. Both of those are limited. Both of them have fairly substantial economic, sorry, environmental externalities. For us, we seek to use atmospheric CO2, the carbon dioxide that's in the atmosphere, as a source of carbon to make our hydrocarbons.
00:05:42
Speaker
which is actually how plants grow as well. Plants derive the carbon in their bodies from the air as well. And so we have a process to filter the CO2 out of the air and then convert that using a chemical reaction back into the fuel from whence it most likely came sometime in the last 50 or 100 years of frenetic industrialized hydrocarbon consumption.
00:06:05
Speaker
Great. So in a kind of nutshell, the ingredients are water, which gets electrolyzed to create hydrogen. And I guess oxygen is obviously a byproduct as well. That's right. And then you're also taking carbon dioxide from the air, which is there in
00:06:26
Speaker
pretty low concentrations, although concentrations have gone up appreciably over the last 300 years and have caused lots of problems. Extracting the carbon dioxide from the air, combining with that hydrogen, and you end up with methane and, I guess, some oxygen via the electrolysis and a little bit more water, right? Is that the kind of... Some water comes out as well, yeah.
00:06:53
Speaker
The process is actually quite a bit more complicated than this, but we have numerous blog posts and so on available on our website and on our blog blog.terraformindustries.com. If you're interested listeners who are deeply into chemistry, they can delve into the finer details there. The really neat thing as far as I'm concerned is that
00:07:12
Speaker
This process can be almost completely self-contained. As you say, CO2 is relatively scarce in the air. It's about one pound in 2000. Actually, if gold was present in rocks at one pound in 2000, that would be considered an exceptionally rich source of gold.
00:07:31
Speaker
It's kind of intermediate in its value. We do have to filter a lot of air to get the CO2 that we need out, but it can be done. This is a process that's routinely done in submarines, it's done in spacecraft, it's done in medical applications, it's done in certain kinds of scuba diving.
00:07:46
Speaker
kits as well to filter the CO2 out of, out of, out of air streams that humans depend on. And actually kind of one of the neat things about processing such huge volumes of air is that if we need a source of water, we can also obtain that via direct condensation. So we envisage, you know, eventually trillions of acres of solar panels and our chemical plants deployed in very remote regions without infrastructure, without
00:08:10
Speaker
often adequate water supplies even underground and in such cases we can actually not only generate a stream of fuel which can be transported to market via pipeline or via trucks but we can also generate a modest stream of water both to supply our process and also potentially to supply local communities or
00:08:28
Speaker
or conduct some limited irrigation on site, which is kind of a neat trick. And all of this is possible because it's driven by solar power.
Deploying Solar Factories in Remote Areas
00:08:37
Speaker
And of course, plants are driven by solar power too. So in some senses, we're doing what plants do. But because we have access to metals and silicon and advanced technology, we're able to produce roughly 1,000 times more fuel per unit area of land than plants can, which is a pretty neat trick. Yeah, indeed.
00:08:59
Speaker
And that's interesting about the water coming off as a byproduct. I've seen that you have an interest in desalination and other means of creating fresh water. But I've not seen you write anything about this, although it struck me that maybe it was on your mind.
00:09:25
Speaker
You know, because carbon dioxide is relatively low concentration, as you said, about 400 parts per million, you have to sift through multivariate, as you say, to concentrate the sift. That gives you the opportunity to also extract a lot of water from water vapor. I guess even in places which are very dry, very low humidity, that can still, even at 5% or lower,
00:09:53
Speaker
That's a lot of water in the air compared to the concentration of carbon dioxide, rather. OK, that's cool. So is this something, then, that could be deployed pretty much anywhere? Or do you need, let's see.
00:10:15
Speaker
do you need water to get it going or could you just put it down in the desert and one of your kind of mini factories as it were that's producing fuel from the air would just get going.
00:10:28
Speaker
I think, practically speaking, you would want to deploy them by the hundreds or thousands at a time, just to make sure you have a meaningfully large flow of fuel coming out. Even though we can produce 1,000 times more fuel per area than plants, that fuel is produced continuously, whereas plants typically harvest it once per season, like crops harvest it once per season. So there's kind of a consideration there. And then obviously, like,
00:10:57
Speaker
The primary constraint on the land is that it's available and cheap. But yeah, it's neat. You could imagine air dropping one of these in the middle of Nevada somewhere, provided that you can get a four-wheel drive road to the site, then that's basically all the
00:11:14
Speaker
all the infrastructure that's needed to make one of these work, which I think is really neat. Uh, the really transformative thing about what we're doing here is that not only can we do this, we think we can do this cheaper than drilling a hole in the ground. So, so maybe not today, but within say 10 years time, even without subsidies, uh, anywhere on earth, even if you happen to be sitting on top of a very rich supply of oil and gas, you know, some parts of Texas or Middle East or whatever, it will probably be cheaper to make hydrocarbons using our process deploying on the surface.
00:11:43
Speaker
without drilling any holes in the ground, without getting any fossil carbon out of the ground and putting the atmosphere, then it will be to go on drilling. And drilling is a difficult and in many ways dangerous business. It's quite high risk. It's also financially risky in that an awful lot of dry wells get drilled over the years.
00:12:02
Speaker
Whereas our process is relatively straightforward, there's very high reliability is guaranteed out of the gate.
Global Economic Impact of Hydrocarbon Access
00:12:09
Speaker
And yet there are associated hassles with operating on the surface and you got to deal with dust and vegetation and so on. It actually seems to us quite likely that if we cover sufficiently large swaths of the world's arid areas and solar panels, that it will probably modify the local climate a little bit, probably make it slightly
00:12:26
Speaker
slightly more conducive to plants growing, which is fine as far as I'm concerned. But it's an important thing. But one of the things that I'm deeply concerned about in terms of human progress is that for the last 50 years, which is essentially most of the lifetime of most of the people who are alive today, humanity has had to deal with fundamentally scarce supply of hydrocarbons. And for people like you and I who live at the very top of the pyramid,
00:12:53
Speaker
It's not a major constraint for us. But for the vast majority of humanity, even obtaining access to hydrocarbons is not necessarily straightforward, certainly in the quantities and at the costs necessary to do productive economic development. And really the cost of hydrocarbons for useful applications in mechanized farming or transportation with personal cars or aviation is really cost prohibitive.
00:13:20
Speaker
And there are a number of reasons for those. Obviously, some countries lost the geopolitical lottery. They don't really have access to seaports. They don't really have the economic development. They don't really have their own oil and gas, which is not great for them. But it turns out that the vast majority of the world's population
00:13:38
Speaker
lives essentially within a day's walk of a place where they could in principle grow enough food to live. So until the advent of refrigeration, essentially all the world cities had to have nearby agriculturally productive hinterlands that could produce enough food to supply that city. And the human population distribution has not changed that much since the invention of refrigeration in part because
00:14:03
Speaker
Inversion refrigeration came along as part of industrialization, which at the same time also caused a collapse in birth rates. So it just so happens that 98% of the world's population essentially live somewhere where it's sunny enough to synthesize hydrocarbons with a solar synthetic process like the one that we are developing.
00:14:21
Speaker
which is super compelling, right? So like now if you're a human being that just happens to be born in Africa or Central Asia or South Asia or South America or whatever, and you live in a country which does not have adequate political organization to ensure reliable supply of energy via what is considered conventional means today, like having a supertaker show up at your highly developed port or whatever, it's not a problem.
00:14:50
Speaker
you can essentially locally finance and construct solar array and synthesize natural gas, you can synthesize gasoline, you can synthesize various kinds of plastics, whatever you may need within your own community or within your own city, which essentially
00:15:06
Speaker
kind of removes that layer of like global competency and coordination from your ability to get enough energy to make your life not suck. I think this is super exciting. And actually when you look at the 2% of humans who don't live in places where this will work, it's basically Finland. So I always feel a bit sorry for Finland. They gave the world Linux and in return they get no solar synthetic energy. But actually there's copious supplies of wind energy and so on at that final as well. So I think
00:15:34
Speaker
If we don't really screw this up by the early 2040s, the vast majority of the world's population will be able to derive cheaper fuel from synthetic processes that are carbon neutral. Right. Yeah, that's an interesting point. So the sun is a great leveler here, instead of you having to win the hydrocarbon lottery and find yourself standing on an oil field or a gas field or what have you.
00:15:58
Speaker
you just need exactly the same resources that you need to, that the plants need. Because as you point out, this is actually chemically not a million miles away from what plants do. It's a lot simpler. It's a lot more efficient. But it's taking ingredients from the air, carbon dioxide, water vapor, and creating hydrocarbons. That's right. At sunshine, obviously. Yes. I actually live in one of the countries.
00:16:27
Speaker
I live in one of the places, not quite Finland, but Scotland. I'm always struck by how much of an outlier it is on the kind of sunshine lottery as it were. We're really low on that list. Well, Scotland's an interesting one because the population density in Scotland is still pretty low in part because my ancestors left the place and went to Australia and obviously because of the clearances.
00:16:53
Speaker
In principle, you could probably pave the moors with solar panels and produce enough hydrocarbons to get by. In practice, I suspect that the British Isles will do better via wind power and nuclear power, and possibly leveraging their blue water navy to import hydrocarbons from other places.
Alternative Energy Strategies for Low-Sunlight Regions
00:17:10
Speaker
But it's kind of a special case.
00:17:14
Speaker
Yeah, basically like Britain, Denmark, and the rest of the Nordics are kind of a special case in this instance. But it's important to remember that Scotland is 56, 57 degrees north or something. It's like at most other places on the planet that are that far north are essentially very, very inhospitable climates. Yeah.
00:17:34
Speaker
Yeah, like we're talking like Anchorage or something like, uh, it's just not, not, not particularly pleasant. And, and the human population density is quite low. Britain is kind of an exception because of, because of the Gulf Stream it's warm enough. Um, and it does, it is sunny enough for like enough grass to grow to like feed, feed animals and, uh, and have some agriculture. Um,
00:17:55
Speaker
So it's an issue on, I was looking, actually this, this might interest your listeners, um, for a while with a couple of my friends, we were looking at like, what would it take for wind to compete with where solar is going to be in 10 years or 20 years? And, and I, you know, I'm, I'm as thrilled as anyone by the incredible developments that are going on with offshore, particularly offshore wind power. And I think it's, it's actually particularly illustrative that, uh, despite the enormous productivity of the North sea oil, uh,
00:18:21
Speaker
like oil area, I guess, oil exploitation area, which, you know, something that was all happening when I was a kid, it's now kind of the situation that if you were able to mount a sufficiently large wind turbine on, you know, the remaining oil platforms out there, that they would generate more energy
00:18:44
Speaker
per decade or more energy per dollar invested than further oil exploitation would. So this is an interesting kind of transition point where if you've got a huge pile of money that you want to invest in generating energy that involves screwing around with marine access in the North Sea, are you better off just doing wind power than you are doing oil drilling, even though the North Sea is relatively shallow? Yeah. And the same thing goes with coal, actually. So actually, I grew up in a place that's pretty close to Newcastle in Australia.
00:19:12
Speaker
which is one of the world's major coal mining areas just as Newcastle was in Britain. But it's kind of got to the point now where unless you find a 10-foot thick, a 3-meter thick seam of coal at the surface, which you're not going to do because all the good stuff's long gone, even then you'd be better off just using it as a foundation for solar panels. You would make more energy and more revenue over a 20 or 30-year time span doing that than you would by digging the coal out and selling it.
00:19:40
Speaker
which is also very encouraging. Yeah, yeah. I mean, I think that's fascinating. And I guess, I mean, this seems almost central to your thesis that solar power is becoming so cheap that instead of it being used for the sort of, I guess, low entropy forms of energy like electricity, which is sort of seen as the most high value forms,
00:20:09
Speaker
it starts to make sense to do something which may seem very counterintuitive, which is actually turn that into, in some ways, less valuable forms of energy, at least from some kind of thermodynamics accounting, at the cost, obviously, of also losing some of that energy in this process.
00:20:35
Speaker
What I'm going with this is I think there's a dominant theme in sort of net zero transition that electrification is the way to go because it's the way that you hold on to as much of the energy that you're producing via renewables as possible. But you're saying, well, you know, let's not electrify, let's produce hydrocarbons in a
00:21:03
Speaker
carbon neutral way if you like, so none of the carbon that you're putting into those fuels is coming from the crust, it's all coming from the air.
Synthetic Fuel vs. Electrification for Aviation
00:21:15
Speaker
Maybe walk us through why that makes sense.
00:21:21
Speaker
Yeah, sure. So it does seem kind of crazy. And actually, I'm very pro electrification. And I think that there are certain applications where electrification is already very competitive. And I expect that its lead will widen. So we've already seen electrification long ago of mechanical mode of power. So it's very unusual now to have a steam boiler driving a turbine in a factory in order to get shaft work to operate a lathe or something like that. You just have an electric motor.
00:21:50
Speaker
And what we've seen over time is that the technology that allows you to electrify things has kind of improved. So electric heat pumps are now significantly better, not just more efficient, but better over to overall cost value than heating in homes, for heating in homes. We're seeing electric cooktops, induction cooktops,
00:22:12
Speaker
have significant advantages, particularly in terms of indoor air pollution in homes. And then obviously, we're seeing the gradual electrification of ground transportation. So I drive an electric car, which is a fabulous vehicle. It is in essentially every way superior to an internal combustion engine version of the same car. And I think that Tesla's explosive growth against a backdrop of general stagnation in automotive is an apt illustration of this fundamental truth.
00:22:40
Speaker
I would expect that trend to continue over time. In fact, in order that solar power gets cheap enough that synthetic fuels can be competitive with fossil fuels, electricity must be so cheap that it will continue to drive the gradual electrification of essentially heating, cooking, and ground transportation. However,
00:23:04
Speaker
It is also important to realize that electrification really kind of stumbles when it comes up against two major problems. One of them is quite hard to decarbonize sectors, and I'll come back to that.
00:23:17
Speaker
There's just a lot of inertia in certain applications. So whereas the fleet of cars may turn over every 10 or 20 years, the fleet of buildings might only turn over every 100 years. And when I was in college, there were still buildings that were heated by coal because it was just a significant hassle to change how the building was heated. There wasn't a strong argument in favor of doing that for better and for worse. And the building was seen as valuable in other reasons. So it wasn't just kind of pulled down and replaced.
00:23:48
Speaker
So for those applications, for legacy applications and so on, I think it's necessary to have a deep source of fuel. But then in the hard to decarbonize sectors, there are certainly electrically intensive ways to process materials, say metals, so making steel, green steel is a good example. And there are also, I think, really transformative ways in which cheap electricity will make, say, aluminium, I guess you're Scottish, I'll say aluminium, competitive with
00:24:16
Speaker
uh other say um more competitive against steel so so right now for example worldwide we produce roughly 20 times more steel than aluminium but uh in the future with you know electricity comes down in price by factor five or so um then then the relative cost advantage of uh steel and aluminium will shift in aluminium for a lot of
00:24:35
Speaker
applications, which is super cool. Also magnesium, I would expect to become much more widely used, and other metals that are extracted electrolytically. And actually, there's also an electrolytic process for extracting iron, which is super cool. So you could
00:24:54
Speaker
If you want to greenify the production of steel, you could obviously replace blast furnaces with synthetic coke. You could fuel the blast furnaces with synthetic natural gas or with ordinary natural gas that you pyrolyze and then install the CO2 or ordinary natural gas and fire it with pure oxygen and then capture a pure CO2 waste stream. Or you could synthesize hydrogen on site and then use hydrogen as a source of heat because electric resistive heating tends to run out of oomph at about 1,100 or 1,200 Celsius, which is not hot enough to make steel.
00:25:24
Speaker
Or you could just go for a direct electrolytic steel production process. But all these different things kind of ignore the fact that existing steel production plants have a lot of fixed costs that's already in place. And so the operators of those plants probably don't want to invest a whole bunch of money in ripping out stuff that already works to replace it with stuff that might not work. So that's kind of a challenge there. And so I think in order to service those needs, it makes sense to have synthetic fuel available.
00:25:53
Speaker
And then the, the most exciting example that I usually bring forward is aviation. So like the shipping as well, obviously, but, um, but aviation is a good example where like, uh, especially long distance, uh, flights. So it's kind of two schools of thought here, right? If we want to decarbonize aviation, one is you make everyone feel really bad about aviation and you invest heavily in alternative to aviation, like, um, sailing ships and high-speed trains, uh, which are actually not significantly more efficient on a per person, like a kind of power per person per mile per time saved basis.
00:26:25
Speaker
or you slap a carbon tax on the price of aviation to help offset the environmental damage that aviation causes because it's extremely fuel intensive. The effect of that carbon tax, of course, is to increase the price of aviation and make it more exclusive and essentially deprive some millions of people in the world from access to
00:26:47
Speaker
the technology of aviation, which essentially allows them to exchange money for time. And at the end of the day, people can always make more money, but they will certainly die one day unless we solve aging. So time is precious. And I think from a human progress point of view, we would
00:27:04
Speaker
probably like to find ways to make flying more accessible to people rather than less accessible. And one of the ways that you should do that is by not making it more expensive. And so the other school of thought, which is the synthetic fuel school of thought says, well, what if we could make sustainable aviation
00:27:21
Speaker
I'm going to take five here. Sustainable aviation fuel, SAF. That's why people call it SAF. It's a bit of a tongue twister. What if we could make SAF that was not only carbon neutral, but also cheaper and a better fuel. They say higher energy density than existing kerosene or jet fuel.
00:27:42
Speaker
and just produce that all over the world wherever it was needed and thereby produce the cost of aviation in addition to making it carbon neutral. Would that be cool? I think that would be pretty neat. Why would that be neat? Well, first of all, with cheaper fuel, we can restart the era of supersonic aviation.
00:28:01
Speaker
You, me and the other 10 million or so people on earth who fly routinely will be able to afford to fly supersonically, which is even more time saved. So instead of taking 14 hours to fly to Australia once or twice a year, I can take four hours to fly to Australia 10 times a year. Okay. Now I'm consuming 60 times more fuel than before.
00:28:18
Speaker
There just isn't enough oil and gas out there to make this work. There's no way that fossil oil and gas can get us to a future where the top of the pyramid, economically speaking, people who are pretty fortunate, you and I, can afford to fly supersonic aircraft around the world frequently, let alone the next 100 million or billion people. But with sustainable aviation fuel made from a solar synthetic basis, then there's no real reason why this cannot occur.
00:28:46
Speaker
And actually, well, this might be getting a little bit off track here, but screw it. In pre-industrial era, so actually the Industrial Revolution in many ways started in, well, I guess it started in South England, but it's certainly a lot of a decode in Scotland. But prior to that point, talking 1750 or so,
00:29:08
Speaker
a good approximation that the total mechanical output that a human had available to them was essentially the amount of muscular force they could bring to bear personally as a result of having consumed some food. So essentially the gross energy consumption on a per capita basis was roughly equivalent to how much food they could find to eat, which in bad years would be around 1500 calories a day.
00:29:28
Speaker
So, actually starvation was not that uncommon. Nowadays, it's more than a hundred to one, right? So like, roughly speaking, I might consume 2000 calories a day, but the machines in the car that I drive and the air conditioners that I use and the planes that I fly on average consume 200,000 calories of energy per
Future Energy Consumption Trends
00:29:45
Speaker
day. And the really mind boggling thing is that the energy they consume is about a hundred times cheaper than the food that I eat on a per energy basis. So while it cost me five bucks to buy a Big Mac, I can buy
00:29:55
Speaker
You know for five dollars I can buy about a hundred times more usable energy than the Big Mac contains in the form of electricity Or in the form of gasoline, which is one of the reasons why? You know people with relatively modest means can drive cars around even though those cars are no way a ton and move significantly faster than Usain Bolt at top speed
00:30:19
Speaker
and yet still be able to afford it. If instead you were driving an animal around at that speed, like, I don't know, a giant mutant crossed cheetah, an elephant, something the size of an elephant around at the speed of a cheetah, feeding that animal would
00:30:35
Speaker
would be extremely expensive, just in terms of the sheer caloric needs that it would have. I guess that the bottom line of this rather rambly sidetrack thing is that today in the industrialized countries, roughly speaking, humans consume 100 times more energy in the form of
00:30:56
Speaker
external to their digestive tract than the energy that they consume by eating food. And ultimately, I think the story of human progress takes us further in that direction. So under severe duress, you were forced to think of ways to increase that number from 100 to 1000 so that people consumed 1000 times as much energy than they consumed as food. There are a very limited number of ways of doing that.
00:31:19
Speaker
Simply purchasing more stuff won't get you there. Simply turning up your thermostat in winter and turning it down in summer won't get you there. Simply living in a slightly larger house won't get you there. Having a bunch more children won't get you there in order to consume 10 times more energy than you do right now.
00:31:33
Speaker
you really have to move really heavy things at very high speeds. Or potentially, this one's a bit strange because it's not quite sure if it should be counted on a per capita basis, but essentially the two major growth areas for energy consumption that I see in the future are aviation and computation.
00:31:52
Speaker
Interesting. Yeah, I thought you might say interplanetary travel for the second piece. So if you take on a per capita basis for the people who are actually involved in it, right? So on a per capita basis for the astronauts, obviously, evidently, it is a colossal sum of energy. And even if you take it on a per capita basis for the people who work in that industry, it's still a substantial increase in the amount of fuel, like any amount of energy that is consumed overall.
00:32:22
Speaker
If we say right now, aviation's adjustable market is maybe a hundred times larger than the current market is, if that makes any sense. If you think of the total number of flights that people go on,
00:32:39
Speaker
And you compare that to the total number of flights that could be gone on if all 8 billion people on Earth flew planes as often as we did or whenever the hell they wanted to. It would be quite different. Just to give you some rough numbers, right now there's like on the order of tens of thousands of commercial flights in the United States every day. So like even in your wildest dreams, how long would it take before there were tens of thousands of Starship launches per day? And how many people would be involved in that? So it just seems like it's kind of,
00:33:05
Speaker
In our wildest dreams, it'd be anything other than a footnote in terms of total energy consumption. I think that's right. And I think also it's the case that we tend to do a lot of short trips rather than, there's some sort of probably exponential decay to the trips that we do. So I reckon even when there's quite a lot of interplanetary travel, there'll be a lot more planetary travel.
00:33:32
Speaker
That will always be the case. I would hope so. I would hope so. I'd say the other thing is for aviation, the cost of the fuel is a pretty significant fraction of the cost of the ticket itself, something between 15% and 30%, whereas for rockets, it's not the case at all right now. In the future, it might be. But right now, that's not the case.
00:33:56
Speaker
OK, this was a super interesting, I guess, set of tangents. I want to get back to, let me see. The kind of original idea was, OK, so electrification isn't the whole story, right? Because there's a lot of legacy hardware and the gas network, for instance. And we just got really prosaic here. The UK has a really good gas distribution network or natural gas distribution network.
00:34:27
Speaker
Ripping that all up, replacing it with a better power grid is not an easy project. It involves going to lots of people homes, changing boilers, changing cookers. And those things will happen naturally as electricity becomes cheaper. But there's no kind of...
00:34:55
Speaker
It's going to take some time. Yeah, I think distribution networks that go to residential areas and so on will gradually be disconnected as the homes in those areas switch over to electrification. And that will potentially require additional investment in local electricity distribution or local battery storage or whatever.
00:35:16
Speaker
But that's kind of in some ways separate to the central question of like, what is it that we're going to be using hydrocarbons for? And I think mostly we'll be using it for industry and transportation. Right. Yeah. OK, so that's a smaller piece, Heating People's Homes. And I guess the biggest piece is, I guess, increasing the GDP of the world by
00:35:40
Speaker
just supercharging everything that we're already doing by creating cheaper hydrocarbons. And better hydrocarbons, as you say, for aviation. We could have ones which are much denser in energy. Well, not much denser, but maybe like 10% denser or something like that. OK. So apparently,
00:36:07
Speaker
Japanese jet A fuel is significantly less dense than American to the point that even though flying from Japan to United States is with the prevailing winds, like Gulfstream jets and stuff will often have to refuel halfway rather than flying the whole way back to the United States, which I think is kind of interesting. But obviously if it's synthesizing fuel, we can make whatever the hell we want.
00:36:30
Speaker
In terms of GDP, we can look at a chart of world GDP as a function of time and it's very clear that something went horribly wrong in the early 1970s when essentially the OPEC countries re-nationalized oil supply and then around the same time production became scarce and cost became less predictable.
00:36:53
Speaker
And I think, I think if we can re secure supplies of hydrocarbons for, you know, 8 billion people on earth, it's a fairly conservative estimate that we can double, um, you know, GDP growth from about two and a half percent to about 5% per year. And just that in itself, like is very cumulative over time. You know, after 10, 20, 30 years, like we end up twice as rich as we otherwise would have been. And that's, you know, that's worth fighting for.
00:37:21
Speaker
I guess one thing that some folks may be thinking here is, OK, so yeah, agree that we want to create new fuels for industry that are carbon neutral.
Challenges and Critique of Hydrogen as Fuel
00:37:35
Speaker
Why don't we just use hydrogen, right? And I feel like hydrogen has received a lot more press than synthetic fuels.
00:37:48
Speaker
By synthetic fuels, I mean hydrocarbons, therefore there is not just hydrogen. Even though one might think of hydrogen as a synthetic fuel when it's produced by electrolysis, I guess it doesn't fit into that bucket. So what's the problem with hydrogen? Why can't we just use that for our smelting processes? Why can't we have hydrogen power planes, et cetera?
00:38:12
Speaker
It's super energy. It's got a lot of energy per kilogram. It's great, right? Per kilogram, yes. Per volume, maybe not so much. So hydrogen has a number of issues. I think there are certainly applications where hydrogen makes a lot of sense and it will make sense to use electricity to generate hydrogen locally in chemical plants and then use that hydrogen to do stuff, whether it's ammonia or whatever.
00:38:40
Speaker
However, hydrogen as generic fuel has a number of severe drawbacks, which I think are not adequately de-risked for it to be competitive going forward. They include basically safety related stuff. The fact that it's not really compatible with existing skill sets or supply chain for standard gas interfaces, whereas like in the past we switched from coal gas to natural gas and they were pretty much cost compatible.
00:39:06
Speaker
And just hydrogen is kind of finicky to work with. I mean, NASA uses hydrogen for the SLS rocket and a handful of other rockets as well. But SpaceX decided not to. SpaceX has some of the sharpest engineers on Earth. They certainly could if they wanted to, but they decided not to because it wasn't worth it. It was not worth it for SpaceX to use
00:39:26
Speaker
Hydrogen is a rocket fuel where hydrogen actually has some limited advantages. Like why do you think it would work for cars or planes? So the major drawback with planes is the density rather than the specific energy or whatever. So like energy per kilogram is fine, but actually energy per kilogram is meaningless unless you include the balance of the system needed to handle it. And so if you end up in a situation where sure the hydrogen is nothing, but the tank required to hold it weighs 30 times as much, well,
00:39:53
Speaker
You haven't got anywhere, right? And hydrogen by itself is even liquefied is more than 10 times lighter than water. So it's just, it's a packaging problem. And in fact, there is actually a pilot hydrogen aircraft, which has been doing the rounds in Europe recently. And I think it has like a 15 minute flight time or something. Maybe that's wrong and someone's going to send me an angry email, but like, certainly it's not going to be able to compete with Transiogenic
00:40:22
Speaker
Um, flights on, you know, seven, seven, seven, seven, so seven, eight, sevens or eight three fifties or eight three eighties or anything like that anytime soon. Whereas as soon as you throw a little carbon atom in there and turn it into even liquefied natural gas, let alone jet fuel or something, uh, the energy density, like the energy per per unit fuel tank, uh, volume goes up by a factor of 10. And like nature is trying to tell us something here.
00:40:47
Speaker
Hydrogen, the advantage of carbon, it's obviously a chemical precursor, so hydrogen is pretty useless by itself if you want to make chemicals, if you want to make plastics and so on. But actually, aviation hydrogen is useful if you're doing a zeppelin. So as a lifting gas, I think if you're building an airship, hydrogen is the way to go.
00:41:11
Speaker
But just as a gas by itself, it has a number of nasty properties. For those of your listeners who are less familiar, it has a tendency to leak through things because it's such a tiny molecule.
00:41:20
Speaker
Uh, the, the tolerancing of valves and so on required to deal with it is easily, you know, 10 to a hundred times more fine than other gases, which is kind of bizarre. It will leak through metals, uh, in doing so it also makes them more brittle. Um, so it affects the mechanical properties. Um, it's very light, so it tends to float away and collect in odd places. Um, it's odorless. So you can't smell it. You don't know if there's been a leak. Um, it has a, what's called, uh, adiabatic free flame velocity. That's about 10 times higher than ordinary.
00:41:51
Speaker
natural gas. So its behavior as a flammable gas is quite unintuitive. It also burns basically invisibly. So unless it's dark, it's very, very difficult to see a hydrogen flame. So if you have a leak that has caught fire, the first sign that you may have, and it burns extraordinarily hot, but the first sign you may have that you have a fire is that stuff starts melting or catching fire around it. And that's obviously extremely worrying.
00:42:21
Speaker
So all in all, it's, uh, it's actually, you know, I've, I've been making hydrogen by electrolysis in personal projects and stuff now for many years. Um, and even at terraform, we've, we've made a lot of hydrogen. We've had one minor mishap, um, with, with hydrogen where we inadvertently melted Bunsen burner. Um, but, um, it's, um, it's, it's a chemical that demands respect, you know, in a way that, you know, like, I feel like we've made gasoline safe enough that randoms can pull up to a gas station and fill up the car, you know,
00:42:51
Speaker
And we've made lithium ion batteries safe enough that randoms can walk around with a phone in their pocket getting dropped on the ground and, you know, uh, accidents that involve injury are rare enough that they're remarked upon, right? Uh, we've made, um, natural gas safe enough that the gas explosions are really quite rare. Um, and also like you can build a house with, with generic labor, right? But when it comes to hydrogen, if you're trying to build a rocket that uses hydrogen, like,
00:43:19
Speaker
The Rolodex of people who know how to do that and know what they're doing with it is, is short enough. You could probably memorize all their phone numbers. Like it's, it's just, it's like the internet circa 1983. It's like a small group of people and that's a real problem. So, you know, I don't want to bang on about this forever, but I think it's important because hydrogen has received like a lot of press, a lot of investment. Um, and I, you know, it's quite striking really. Um,
00:43:47
Speaker
I think one thing you didn't mention is that the ignition energy necessary to blow up some hydrogen is really low, right? You mentioned the cell phone falling to the ground. That might be enough, right? The spark of something on your cell phone. So if you have a mixture of oxygen and hydrogen, the amount of energy required to ignite it, I think, is 0.02 milligels, which is
00:44:17
Speaker
In principle, a fly landing can dissipate that much energy. And the other scary thing about hydrogen is that it has kind of unpredictable, first of all, it's exceptionally flammable. So across a wide range of mixtures with air, it will catch fire, whereas most other gases have a much narrower range of mixtures where it can burn.
00:44:46
Speaker
And then it can do this thing called DDT, or Deflagration Detonation Transition, where you have a hydrogen fire, and then for basically no reason at all, it just detonates. And you end up with a supersonic combustion front, which is obviously vastly more damaging. Yeah, so all these things are kind of unpleasant, to put them hardly.
00:45:12
Speaker
Yeah. So it's hard to store. It's not very useful as a fuel because it's so light. All these problems can be solved. All these problems can be solved. It can be handled safely. It can be dealt with safely, et cetera, et cetera. But they all cost money. And so once you end up with a system that's safe enough that it's safe enough for ordinary consumers to use, it's not even remotely competitive on price. And it never will be.
00:45:42
Speaker
And, and more to the point, it will never be able to out-compete existing hydrocarbons. Right. And it won't be able to out-compete synthetic hydrocarbons and it has already lost like hydrogen fuel cell cars had already lost to battery electric cars in like 2006. So I just don't know what, like I'm going to call that Toyota here. Like Toyota, like was the absolute pinnacle of automotive manufacturing for after, after many, many decades of extremely hard work, they got there and they held it for decades. They held it for several decades, right?
00:46:12
Speaker
And they had an early investment as did Daimler in Tesla. Right. And, and their engineers who I've met, you know, they used to come to Caltech and try and like recruit people knew as well as anyone else that electric battery electric cars already in 2008 had better performance than hydrogen fuel cell cars could ever achieve, even if they achieved 100% efficiency, et cetera, et cetera, which they were never going to. And yet for some reason.
00:46:39
Speaker
Toyota went and doubled down on, on hydrogen sold their stake in Tesla. Like unbelievable, absolutely unbelievable. Um,
00:46:47
Speaker
Anyway, I don't want to pick on Britain because Britain's having a fair share of similar mistakes. Maybe it's just a property shed by, uh, by islands that lay off the coast of Eurasia. Um, but, uh, I think making mistakes is a good thing, but, uh, yeah, you have to know when, when it's time to throw in the towel and move on to make another mistake. Right. Um, yeah, I guess it's hard to believe that Toyota could throw away its lead that easily. It's just, yeah, it's shocking. Anyway.
00:47:15
Speaker
So I guess, I was just gonna say, yeah, yeah, I think it was full cells was Elon Musk's term for fuel cells, which was a nice piece of empty marketing. But yeah, so I guess what you're doing is taking the, you do have hydrogen as part of the process, but you combine it very quickly with the carbon that you got from scrubbing the air and that gives something much safer, much easier to handle.
00:47:44
Speaker
people know how to handle it. We have the pipes to do so all over the place. Like, um, we have the expertise to use it. You can produce plastics from it. Um, yeah, you can, you can burn it in your cases if you want to, although as you point out, like it's, uh, it's incredibly versatile, right? And I don't really care what people use it for. Right. I provided that invented directly into the atmosphere out of spite, uh, which makes, uh, you know, methane is not a great greenhouse gas. Um, yeah, or rather it is a good greenhouse gas, which is not a good thing. Um, as long as they use it somehow.
00:48:14
Speaker
But the thing is, I don't have to specify how they use it. And part of the problem with the hydrogen economy is that the development is still stuck in the ideation phase. How might we do this, that, and the other? And it basically comes down to a live action role play exercise in redesigning the entire industrial economy from the ground up, just for fun. There's no real economic reason to do it. And so in that, it shares a certain commonality of spirit with Leninism in that it presupposes that a centrally designed economy
00:48:44
Speaker
Could ever possibly compete usefully with with a decentralized one and and that there it kind of betrays It's the weakness of the fundamental weakness of the idea as you say The ironic thing is that is that when synthetic fuel plants if they're using our system or similar up and running Not only will they be generating vastly more hydrogen every minute than humanity currently generates every year
00:49:04
Speaker
And not only will something like 80 to 90% of the world's electricity be devoted exclusively to producing hydrogen, all that hydrogen will only last for a few seconds before it's routed into a reactor and swiftly used to tear oxygens off CO2 and then put some more hydrogens onto those carbon atoms and make methane. So it's an interesting thing. I'll put it that way. Yeah, yeah. Yeah, I mean, it's only the case. If we wanted to build
00:49:32
Speaker
If we wanted a lot of fuel cell cars, my goodness, we need to replace a lot of stuff, a lot of infrastructure, and that just doesn't seem so easy. Whereas, for example, if you wanted to run a lot of methane cars, it's super easy, actually.
00:49:48
Speaker
My wife is Argentine and if you've ever looked in the boot of a taxi in Argentina, you'll have seen like a compressed natural gas tank. It's just cheap. Yes. It's the same in Sydney. It was done that way in Sydney. I don't know if it's still the case, but it was very common when I was a kid in Sydney and in Central Asia as well. It's quite, quite normal for cars to operate off compressed natural gas. But actually like, so first of all, I think the cars will mostly go to battery electric in the future as well trucks.
00:50:17
Speaker
If you want to run your classic car that runs on, on diesel or on petrol right now, that's fine. You can synthesize petrol just as easily, really, as, as we synthesize natural gas. I mean, it's like incrementally more difficult, but it's, it's something that's well understood how to do. Um, there are, you know, it's people have done it for more than a hundred years at this point. So it's, it's, it's very doable. Um, and I suspect that that will also get be done in the future as well. That has to be done for aviation. So, um, most likely, um,
00:50:48
Speaker
this makes me I want to pick up on something that you mentioned a minute ago as long as folks use the methane that you're using you don't really mind what they're using it for it as long as it gets used I guess I mean as you say it's a pretty well as long as they pay for it
00:51:02
Speaker
as long as they pay for it as well. It's a fairly nasty greenhouse gas. I mean, it's, uh, I think of order a hundred times worse than carbon dioxide over sort of a couple of decades. Um, so it lasts, it doesn't last nearly as long. So it just naturally oxidized in the atmosphere. Actually. So one of the neat things we can do because we're filtering out so much air, we've looked into putting a catalytic oxidation, um,
00:51:32
Speaker
basically metallic layer, metallic filter layer into our beds so that we'll be pumping large swaths of the Earth's atmosphere through our beds every day. And we can use that process to incidentally catalytically oxidize methane that exists naturally in the atmosphere, which is a neat trick. So we can take down it. So for example, if we end up in a situation where we have massive methane releases from the Arctic because it's thawing out,
00:52:02
Speaker
then the only way we can really do anything about that is by putting catalytic oxidation in the stream, which we're doing, sorry, in the stream of the air that we're processing, which is kind of fun.
00:52:19
Speaker
That is cool. So to envisage the ability to, I mean, if the process is running sort of in its vanilla format, it's completely kind of neutral. Everything that gets produced, at least once that methane is combusted, right, everything goes back to its original place, right? You have as much. Yeah. Yeah. Well, some fraction of hydrocarbons
00:52:44
Speaker
Some fraction of hydrocarbons get converted into plastics and other chemicals, which are durable, durable forms of plastic. And actually, you know, one of the thought experiments that I did early on was like, okay, so how much CO2 do we have to capture, turn into plastic and then pile up in order to keep up with ongoing oil and gas extraction. And it works out to something like you get to build a Mount Everest every couple of years. So.
00:53:05
Speaker
Okay. So like you can make these just absolutely colossal piles of plastic that reach into the stratosphere. And in terms of environmental impact, people don't, people think of plastic as kind of an negative cause like it ends up in bird, bird stomachs and turtles and stuff like that. Um, but actually like if you're just, if you're just making a giant pile of plastic out in the middle of the desert, it is so much better for the planet than having all that CO2 floating around the atmosphere. As long as you don't wash the plastic into the ocean, it's not a big deal. Right. Um,
00:53:32
Speaker
It's just, it's becomes a carbon hybrid items. Yeah. Yeah. Well, so Lego, what's it, ABS, ABS has other chemicals in it, I think. Yeah. Yeah.
00:53:46
Speaker
No, no, it's a pure hydrocarbon. Okay. Acrylonitrile. Oh no, it has a nitrogen in there, but, but yeah, I mean, you could, you could make plastic out of it. We could take a little bit of nitrogen out of the air. You could make giant Legos. You could make Legos, like Legos the size of aircraft carriers and plug them together. I like the size of that. The sound of that. I think people would be very happy with the price of Lego coming down as well. Like this, this should be in your pitch deck. Lego has become, it's already so much cheaper than when I was a kid.
00:54:17
Speaker
Really? Yeah. Like to me, I buy like Lego rockets and things. And for me, the limiting factor is the time it takes me to assemble them. Whereas when I was a kid, obviously limiting factor was like, my parents are too cheap to buy Lego. Um, but I lived in Australia and you know, the colonies that didn't have access to the supply chain. Um, well, I can see that you're on a mission to change that. Um,
00:54:42
Speaker
So let's come back to one thing I guess I touched on, which is that absolutely key to this all working.
Solar Cost Reductions and Synthetic Fuel Expansion
00:54:50
Speaker
And this is maybe we've glossed over. It's just.
00:54:53
Speaker
the price of or the cost of solar energy coming down. So at the moment there are, I think I've seen in your blog, some places where it would even now be cheaper to produce methane using your process than it is to buy it on the markets, which are, you know, are you getting it from someone who's
00:55:18
Speaker
somewhere extracting it from the ground. And that's a striking fact, but for this to take off internationally, it needs the cost reductions in solar to continue. And I know you're very bullish on that. When do you think this becomes something which would make sense, I don't know, let's say in 50% of the global population or so?
00:55:48
Speaker
If you follow existing trend lines forward, it appears we'll hit cost parity without any subsidies for 50% of the world's population by 2035 or 2036, which is seven or eight years from now. No, hang on. 12 years from now.
00:56:04
Speaker
So, and this is a fairly straightforward calculation that we did months ago because the solar insulation across the world is well mapped. The price of natural gas is reasonably well understood in various markets and population density is also well mapped. So it's not a particularly, actually, I don't want to underrate it. It was a pretty sophisticated calculation, but I did it anyway to understand this.
00:56:28
Speaker
Now, one can kind of draw these extrapolations forward and say, well, look at that. Solar is going to get 10% cheaper every year at that rate. A factor of two in eight years and a factor of two is enough to cover the gap between Arizona and New Hampshire or something like that in terms of solar availability. Or I don't know, like Southern Spain up to Germany or something like that. Southern Spain to Belgium in terms of solar availability is a factor of two. So eight years and you're done. And of course, you might ask, well,
00:56:57
Speaker
Why do we believe that solar will continue to get cheaper? And there's a couple of different reasons for that. First of all, you can take the word of people who are much smarter than me, who spend their entire careers essentially pricing risk on these sorts of developments, and then look at the fact that they've already invested trillions of dollars in expanding solar panel production in the United States to the tune of more than a terawatt of additional solar production
00:57:21
Speaker
capacity by 2027 coming online. And so the global production last year was about 268 gigawatts. So we're looking at like a Forex expansion over global production within the United States borders within three or four years, all of which bakes in certain assumptions on cost that basically get us where we need to go. Okay. But maybe they're wrong. They've been wrong before. Why is it that solar gets cheaper at all? And the major reason there is that as production expands, people have to build new factories and they feed the learning from building the previous factories into the new factories. And so for manufactured technology,
00:57:51
Speaker
not all of them, but many of them follow a thing called Wright's law, which actually first discovered in the context of price reductions in building aircraft in the Second World War. And so depending on the technology, you kind of get a fixed percentage reduction in the cost of additional production per doubling of production. Well, that was kind of a rounded about way of saying it, but I'll rephrase it. Every time you double production of solar panels, the price comes down between 35 and 40%.
00:58:19
Speaker
a factor that's robustly seen with automotive, it's seen in computer chips, it's seen in batteries, it's seen in a whole variety of different technologies, it's seen in aircraft obviously. And depending on the underlying sophistication and complexity of manufacturing and a number of other factors basically determines what that percentage is. But if you back test that, it's been that number for about 50 years now. So like solar has been fairly robustly at 35 to 40%
00:58:45
Speaker
cost reduction per doubling your production now for 40 or 50 years. So why does it do that? Well, as the cost comes down over time, the market, the addressable market expands, right? So demand increases and then that demand increases by an incremental fraction that is more than enough to make up for the expanded production that caused the cost to come down in the first place. And so it feeds back on itself in an expanding way. And what you tend to find over time is that the ability of
00:59:15
Speaker
factories to scale up and meet that demand actually lags, increases of demand over time. And you would think that as the market eventually saturates, that gap would close, but it's not. It's actually widening for solar. And the reason for that is that people thought that global demand for solar would saturate at about 1%. And right now, solar is generating just over 1% of global energy.
00:59:38
Speaker
But actually, it's the other way around. That 1% solar is just barely reaching cost parity in a relatively small number of markets. And its total address market is continuing to expand. And if synthetic fuel is correct, not only will it continue to expand until it's about 100%,
00:59:54
Speaker
of total worldwide electricity consumption will actually grow well past that to more like 2000%, which is to say that we will end up generating significantly more power from solar in say 20 years time than we generate using all sources today by roughly a factor of 10 or 20 in order to convert
01:00:12
Speaker
a large fraction, almost all of that electricity into synthetic fuels and thereby to displace fossil oil and gas production, which right now is not really downstream of electricity production. Now, this is about to get even more complicated. So hang on to your hats, because when solar cost drops by say 10%, the addressable market increases by 20%. But that's not always smoothly the case. And in our case, when solar drops down below about 15 bucks per megawatt hour,
01:00:39
Speaker
or 1.5 cents per kilowatt hour, the addressable market expands from basically some subsection of the electrical grid, which is where solar is addressing right now, to also including synthetic fuels, which is roughly 10 times the size of the grid. So we expect to see, as solar cost drops in various markets around the world, a dramatic expansion in demand.
01:01:08
Speaker
That expansion in demand will translate into a vast increase in revenue and profit for these industries, which will in turn attract vast numbers of extremely clever, highly motivated and hardworking process engineers into factories all over the world in practically every country on earth who will in turn feed innovations in, which will not only improve the learning rate, it will also decrease the doubling time. So Sola's doubling time historically was about three years. It has already decreased to closer to two years and is well on its way to being less than two years as it will have to be.
01:01:38
Speaker
Now, it is incredibly important that we keep that number small so that solar production increases as quickly as possible. Because if we don't, then we could end up in a really messed up situation where solar synthetic fuel has reached cost parity or better all over the world. But there's a 10 year backlog, essentially waiting for solar production to catch up, which means we will go and have an additional 10 years or 500 gigatons of CO2 emission because we cannot somehow figure out how to convert the Earth's crust into solar panels quickly enough.
01:02:08
Speaker
But fortunately,
01:02:11
Speaker
this industry is mature and it appears responsive to capital. And that means that already just the expansion of production, which is on the order of hundreds of billions of dollars now has been enough to kind of smooth that out. And so that's why we're seeing these massive investments in expanded production here in the United States and in Europe and other places. And so I would say compared to a year ago, I'm significantly more confident that Terraform will be able to scale within that umbrella before we saturate
01:02:39
Speaker
the supply of solar panels and essentially limited in our growth by the availability of solar panel production and its growth forever. So we'll have to see how that goes. But I'm reasonably confident now that that will not occur until well into the terawatts of global deployment per year. Okay, I want to play some of that back because I just think it's like a wonderful
01:03:03
Speaker
Yeah, it's a wonderful idea that there's going to be this kind of phase transition where at the moment, you can only add so much solar to the grid, right? We don't have enough battery capacity. We're reducing batteries as fast as we can, but they're not great for long-term storage anyway. They might be in the future. In the future, yeah. Right now, the economics don't really make sense, but
01:03:32
Speaker
You can only saturate your grid with solar, and then it doesn't make sense to buy more. Plus, we already have a lot of energy for the grid, which doesn't make sense to turn off. You can only do so much, but as soon as it becomes cheaper to get your gasoline derived from solar,
01:03:59
Speaker
Wow, the demand just skyrockets, right? You're no longer limited at all. It's kind of like you've dumped over a valley or something, you're in a completely different potential.
01:04:13
Speaker
And I guess your main concern here is, will the solar suppliers be able to keep up with that demand? And sure, there's going to be some lag. They definitely will not be able to. So if you follow the trend of cost reduction as a function of production increase, and then you assume that production can increase infinitely quickly, or essentially as quickly as people can put in your orders, then because
01:04:42
Speaker
because every incremental cost reduction expands the market faster than that production increased, essentially you end up with a runaway chain reaction. And so, you know, when I ran this simulation on a monthly, a kind of a monthly refresh cycle, it was all over in three or four months, which is never going to happen. So like, solid production will never be able to quote unquote keep up, but if we're lucky, we'll only be lagged by a year or two. Yeah. So I guess in some ways, like the learning rate is less important to your,
01:05:12
Speaker
or business than just the rate at which solid production can be increased. And those things are probably related. But I don't know if you have an intuition of how fast that can be. Yeah, there's an argument that you could say, well, can we trade learning rate for increased production? And you could certainly imagine a situation where instead of building one new factory and then feeding its learnings into the next new bigger factory and so on, every step of the way, you just 10x the number of factories you're building.
01:05:41
Speaker
And so you're getting 10 times as many factories per unit of learning. And so that would necessarily slow down the learning rate, at least the first time you did it. But I strongly suspect that learning is incurred as you build the factory. So the amount of learning you get from building 10 factories on the identical plans is more than you get from building one factory, but maybe less than building 10 distinct factories with individual plans or something like that. But yeah, I would certainly,
01:06:09
Speaker
Um, and I think, I think the market will reflect that as well. Right. Like once, once we have a, a solar manufacturing plant that is able to deliver solar panels at the required cost to, you know, then service some swath of the world's population, people will copy paste that factory design and figure out how to make the factory cheaper. Um, so they can make more of the factories faster.
01:06:34
Speaker
rather than making the panels themselves cheaper. That's actually what we've been seeing in the last few years here since the COVID supply chain disruptions. It's been a lot of innovation on factory production rather than the cost of the panels themselves.
01:06:48
Speaker
Yeah, I mean, to some extent, it doesn't matter if the learning rate slows down once you hit that threshold where it becomes cheaper to use your hydrocarbons via solar. There's no going back, right? Great. If solar gets cheaper, brilliant, right? Everything gets cheaper. But I guess for some folks, they might say, OK, I'm a skeptic here.
01:07:13
Speaker
Yes, we've had a learning rate, but maybe we're on some kind of S-curve. I'm trying to play devil's advocate here because, as you say, this has been going on for 50 years, and as long as it doesn't stop in the next few years, let's say the next decade, we're golden.
01:07:32
Speaker
Yeah. Is there something that could stop us? I guess we're maybe not the best people to answer this, but is it just completely inevitable or is there something that we're not looking at? Well, you can play counterfactuals, right? So what if the learning rate dropped to 5% instead of 35%? Well, we'd still get there sooner or later, right?
01:07:51
Speaker
Would the learning rate dropping to 5% cause the markets to expand so slowly that the production rate couldn't be increased anymore? No, it would still increase. There's actually a paper that came out Oxford, I guess late last year, that pointed out that it basically took a brief historical survey of all the reasons why people thought in the past that solar panels could not get any cheaper.
01:08:16
Speaker
and mocked them soundly and profoundly in the kind of Oxfordian accents. So around about 2010, a lot of solar manufacturing occurred in Germany. And German manufacturers thought that they basically run out of ideas. And then solar manufacturing moved to China, and it got 10 times cheaper since then. And that's not because Chinese labor is so cheap. Actually, Chinese labor is more expensive than Mexican labor. So basically, they just took another crack at the problem and solved it.
01:08:45
Speaker
So, and at the same time, solar production increased by, I don't know, a factor of a hundred or something. Um, it's really, um, it's really quite staggering. And if, and I guess the other reason you can look at what a solar panel is fundamentally it's, it's a glorified sheet of paper. Like it's, it's a two dimensional structure made of a chemical that's relatively abundant in the earth's crust using a process that is basically scalable and not particularly sophisticated. The Siemens process has been around for 50 years and in fact,
01:09:15
Speaker
It frustrates me in that it, I feel like it's primitive. It feels like it's like banging rocks together, like kind of level technology. Um, and I feel like our descendants will mock us for making solar panels so stupidly. Um, and if you think about like, what is the essence of a solar panel and how, how would we, if we could arrange atoms arbitrarily, how would we do it to make a solar panel? It would look quite different from the way that we do it right now. Um, and so I think we'll probably trend in the direction of, of solar panels that are even, even thinner and less material and even cheaper and faster to process.
01:09:45
Speaker
Right, yeah. And is there a constraint, do you think, with getting them installed?
Regulatory and Supply Chain Barriers
01:09:52
Speaker
So having all these panels, one thing. Coming back to this point, you mentioned that in some places, so certain areas of Europe, it would be cheaper to run your process already. But that's not happening. And I feel like part of the reason here is
01:10:13
Speaker
Well, it's clearly not the service too expensive, but there are difficulties maybe with a large scale deployments. Um, is that something that needs to change? Yeah. Well, I mean, in terms of the mechanics of deployment, um, there's a lot of potential innovation, potential innovation there, but, uh, I think we've, as a species, barely even pushed on that because it hasn't been a strong economic incentive to do so until very recently.
01:10:40
Speaker
Um, and so I was expected to see maybe a factor of two reduction in total deployment cost and balance of plant cost in the next few years. Um, but that actually the major constraint on solar deployment in Europe right now, um, is, is actually not. Not due to the mechanics of deploying them in the field, it's regulatory and to an extent supply chain. So, so one of the things that frustrates me is like in the aftermath of Russia invading Ukraine, again, um, it seemed to us that we could probably produce a natural gas in Europe.
01:11:09
Speaker
Um, in fact, Europe could produce essentially Europe was in a position where it was, it was paying Russia a billion dollars a day for oil and gas and, and that it could do that, uh, indigenously within its own borders. It cut itself off from Russian gas in just a few years of intensive development with solar panels, uh, of, I can remember how many 30 million acres or something like roughly the area of Belgium of solar panels spread across the European continent. Um, and, um,
01:11:36
Speaker
I never liked Belgium, so I think we could just put them all in Belgium, by the way. Yeah, let's take a vote on it. Half the Belgians will probably agree with you. But the thing is, it's not a huge amount of area. It's probably comparable to the total amount of area that's used to grow citrus in Europe. And Europe's not well known for its deserts.
01:11:57
Speaker
But easily 20% of Europe is essentially unpopulated, non-economically used land. So there's no shortage of land to do it either. And so the major constraint then is essentially Europe would have to stand up major solar panel factories within its own borders. And even now, a year later, there's been no effort to do that. So as far as I can tell, whereas you could imagine
01:12:21
Speaker
Like if they understood the urgency of the situation, like just getting the hell on with it and doing it, as was done, I think with liquefied natural gas importation where there was like a lot of movement there very quickly. And also to an extent with COVID vaccines, you know, obviously there are a lot of frustrations and slowdowns there as well, but ultimately they were delivered pretty quickly.
01:12:42
Speaker
The crazy thing is the total cost of doing that would be reasonably small compared to the ongoing cost of paying Russia hundreds of billions of dollars a year for oil and gas. I found it very frustrating. The other major constraint, aside from supply chain, is regulatory. This is actually a problem we see in the United States as well, which is that, at least in the United States, the major constraint on deploying solar panels now is
01:13:05
Speaker
what's called the grid interconnection queue. So it's about a 10 year process to obtain permission to build a, uh, a solar power plant and then connect it to the grid. Um, and, and that is due mostly to, uh, what's called NEPA, like, uh, environmental protection act, um, processes and also, um, kind of the situation we find ourselves in, in terms of like the legality of, or the legal difficulty of building new
01:13:35
Speaker
power lines that cross other people's land. Essentially, you can end up in a situation where every man and his dog gets to line up and say no, and you have to fight them in court for 10 years to get permission to build a power line, which is, I guess, good for us because our process produces natural gas, which you can just put on a truck and move on the road, and there's no law against that.
01:13:59
Speaker
hydrocarbon industry by and large is not particularly struck like heavily regulated, which is nice. But it is also like kind of a wake up call, I think for us as a species that if we don't get out of our way, we're going to be in serious trouble.
01:14:12
Speaker
And I think the same applies in Europe as well. Like if you, uh, it pains me to have to say this, but if you look at the European response to these, uh, large language model, uh, tools such as chat GPT, um, it's, it's astonishing just how badly, uh, European kind of regulators continue to fumble when it comes to, you know, essentially, uh, leveraging new technology.
01:14:40
Speaker
Yeah. And I guess this is something that really shines through in your strategy is that you don't want to rely on any subsidies. You'll take them if they come. Well, the Inflation Reduction Act is huge for us, absolutely huge. But yes, long term, they cannot work.
01:15:04
Speaker
Like once we're at 1% market saturation, the subsidies will be impossibly expensive to maintain for any state. So we kind of get that little 1% for free in order to scale our processes to the point where we can make it on our own. Yeah. And it seems like the model is geared to
01:15:25
Speaker
work with very small deployments which I suspect would help with getting, you know, you don't need to solve all these problems, right? Your kind of typical unit I think is like a kind of five acre solar farm and then, you know, relatively small bit of machinery that sits in the corner, scrubbing the CO2 out of the air and producing everything else.
01:15:48
Speaker
And your business is not gonna be finding the sites. It's more of a sort of sell the machines, let others figure it out, how to locate them, get around the red tape and so forth. So you're gonna rely on the kind of collective ingenuity of a whole lot of people motivated by the profit that they'll be making on those machines. Is that a kind of first story?
01:16:18
Speaker
Oh, absolutely. Not fighting against the engine of capitalism is rule number one. The key problem with hydrogen fuel cells, as we talked about earlier, is that not only do they suck on a technical level, but no one is going to beat down your door to buy them because they're really expensive. They're just bad value. And that's not the case. If your product is that, like that, you got a problem. But in our case,
01:16:44
Speaker
we believe that our product actually has the ability to deliver huge value for customers. And so it's just, it is just kind of the rule of the market system under which we operate that if you're able to do that, then rivers of capital flow to your door to make it happen. Um, because you know, the market is all about finding ways of getting more money moving and that's, that's what we do. So again, like we can assume, we can safely assume that greed and self-interest will carry the day. And so we want to align ourselves with those forces as much as possible.
01:17:15
Speaker
Yeah. I want to touch on a couple of things I've picked up, which is that you want to make this, you know, these machines, you're aiming for simplicity over efficiency.
Efficiency and Cost-Effectiveness in Energy Production
01:17:26
Speaker
I think you even say rather provocatively that low efficiency is a good thing. And then I think that's kind of leading to the fact that, you know, increases your demand on solar, perhaps, therefore,
01:17:40
Speaker
leads to further reductions in the cost. That's a nice side effect. We're not trying to consume extra solar power. At the end of the day, what matters is the dollar value, the cost of producing fuel. And any particular design feature you'd like to add or take out, you can calculate its effect on the bottom line.
01:18:08
Speaker
And so if you want to add 10% efficiency, you would expect that that would make it cheaper. And if it makes it more expensive, then you should not add it. So efficiency is a good thing except when it's expensive, right? Well, efficiency is a neat thing because it's what's called an intensive property. So you can compare systems of different sizes using the same number. So you can compare your motorcycle and a car and a truck and a plane using the same number efficiency. The way you do it is you measure like
01:18:37
Speaker
the quantity that you want out, the figure of merit, the utility, and you divide it by effectively the cost, which in the case of energy efficiency is the amount of energy you put in. So what is the value you get out divided by the cost you put in? That gives you energy efficiency. But why do we care about the energy we put in? Because energy historically has been a pretty good proxy for cost. Energy is the fundamental factor of production. It's a thing that's hard to come by. It's a thing that we starve for a lack of. But we're actually in the process of building energy. So it doesn't make any sense to divide energy by energy.
01:19:06
Speaker
Or you can, but it may not be particularly illuminating. What people care about is how much energy can I get per dollar? How much stuff can I get per dollar?
01:19:17
Speaker
Energy efficiency is a very good measure of that for certain applications where the application itself is exposed to energy cost in a large way. So a gas-guzzling four-wheel drive, for example, a large fraction of its ongoing costs will be paying for gasoline or petrol to fuel it. And where you expect the cost of that petrol to increase over time.
01:19:42
Speaker
So then you want to say, well, I want to make sure that in the future I'm not paying fast sums of money for the energy to keep my car on the road. So I'd like it to be a relatively efficient one. OK, fine. Whatever. It's all about utility per dollars. But if you expect gasoline to get cheaper in the future, then why the hell wouldn't you go and buy a gas-guzzling four-wheel drive? You'll end up spending less money in the future. And that's the situation we find ourselves in, which is that we expect solar power to get cheaper in the future. And so there's no point in us investing time and effort right now and trying to find ways of doing this efficiently. Like right now,
01:20:11
Speaker
we're very, we're as scarce as we'll ever be in terms of availability of fundamental resources, like, you know, money and time and people and materials and so on. We need to be really careful about what we spend our time and money trying to solve. Is there much point in us trying to solve efficiency? No, why? Because it's not going to be important in the future. And then in terms of exactly how important it has to be, well, we have a spreadsheet for that. And, uh, and you can put all the efficiencies into the top, gives you the dollar value at the bottom. And you can, um, you can basically, uh,
01:20:37
Speaker
Calculated curve of utility is a function of efficiency. And for us, the peak utility is a roughly 50% electrolyzer efficiency, which is still relatively efficient. It's just way less efficient than the state of the art, most efficient electrolyzer possible with known technology, which is about 80% efficient. So suddenly, a reduction of 30% efficiency is not the end of the earth. Early steam trains were less than 1% efficient. Like early steam engines, less than 1% efficient. Later ones pushed 20%, 30% efficient.
01:21:06
Speaker
It's far less drastic than the improvements that thermodynamics wrought in the operation and production of steam engines. I think this is very interesting as it's a bit of a...
01:21:22
Speaker
it separates you from some other folks in this market where perhaps just the engineering mindset is to chase after efficiency. But you're looking at letting the reduction in cost of solar do that work for you in a sense. So instead of you having to extract a higher percentage of useful energy out of the inputs, the inputs just get cheaper.
01:21:51
Speaker
use more of them. And the game is about just building the machines, right? Just building these factories, which are going to run the process as cheaply as possible. And that's getting cheap. Yeah. Yeah. If sold is getting cheap, the question we want to ask ourselves is how do we derive the maximum possible utility from solid getting cheap, right? How do we, how do we leverage that to the, to the, to our best advantage? How do you take advantage of that? Right.
01:22:20
Speaker
And the way you take advantage of cheap energy is by using a lot of it to solve your problem. Like it's not actually more profound than this. Let's take another example. Look at what SpaceX does, right? A lot of industry experts looked at SpaceX and said that what they're doing will never work because they are using the wrong fuels. They should be using hydrogen and oxygen, which are more efficient, right? And there was SpaceX and they had the same engine.
01:22:47
Speaker
for their first and second stage, which ran on RP-1, which is a form of kerosene and oxygen. And what do they do? Why do they do that? Because what matters is the dollar value to the customer of launching something into space. The customer does not care what the fuel is. They don't care what the rocket's made of. They don't care how the rocket was made, that what they care about is being able to accurately estimate the risk of a launch failure, because you can insure against that, and the cost per kilogram of putting
01:23:17
Speaker
their object into space. And what Elon Musk correctly recognized was that the hassle factor of using hydrogen would more than outweigh the cost benefit of using it. So in practice, if he had filled up his rocket with hydrogen, he may have been able to launch incrementally more stuff to space, right? A slightly higher payload, but the launch cost would have been probably four times higher. So for four times the money, you can launch 10% more stuff to space.
01:23:47
Speaker
If you're fundamentally limited by the size of your rocket and you're trying to optimize for how much stuff you can launch per rocket launch or how much stuff you can launch per truckload of rocket, then yes, you have to go to the higher efficiency fuel. That's not a constraint they operated under. And in fact, most of their customers didn't use or use so little of the payload they had available. They're actually able to take their first stage, still full of fuel, fly it back and land it.
01:24:11
Speaker
Right. So they end up with, they still end up with more performance than they needed in most cases. Um, you know, at the end of the day, I think both space X and Tesla, their products embody performing the cost benefit analysis. Like you want to make a design change, show me that the cost for the customer, show me the value for the customer improves. All right. So I can give you another concrete example in the case of the Tesla model three in development.
01:24:39
Speaker
The most expensive component of that car is the battery, right? So if you're able to improve the efficiency, now this is where efficiency is important because it drives us energy from the battery. If you're able to improve the efficiency of the motor system or of the aerodynamic cover of the car or whatever by some fraction, you can translate that into a reduction in size of the battery, which translates directly into a cost reduction. So you can compare the cost of adding a part that improves efficiency
01:25:09
Speaker
with the cost reduction on the battery from that improved efficiency. And if that's positive, then you've made a good decision and you can put that design change in. If it's negative, then maybe you should look at something else you should delete in order to continue along that path until you find the optimum. And so in this case, there was some extruded plastic component that fitted into a hole under the car. I can't remember the exact details. The component itself cost, I think, $0.50, including fabrication and installation.
01:25:37
Speaker
And it improved the error resistance enough that it enabled them to save two bucks and 57 cents on the cost of the battery. So it was a net $2 saving. So it made it in. And the thing is Tesla did this thousands of times on every single subsystem of that car, which is one of the reasons why the Tesla Model 3 and Tesla Model Y, which is basically the same car, are like 50% better
01:26:01
Speaker
50% better value in terms of like performance and range per dollar for the customer than all the other competing cars that basically took a car that looked very similar, put a battery with a very similar size, very similar looking motor, very similar infotainment system seats and so on in there, but didn't bother to turn the screws on every single one of those subsystems to squeeze out the maximum value. And in the end it hurt them to the tune of 50%, which is why Tesla is selling 2 million vehicles this year and their competitors are struggling to sell 10,000.
01:26:30
Speaker
Yeah. Do
Competing with Fracking through Machine Optimization
01:26:32
Speaker
you see this as the model for yourselves to look at the details of everything in the process that you're doing and figure out, here it's not about efficiency, but I guess it's about making it maybe as robust as possible, as cheap as possible to produce these machines. Is that the kind of high level that's the game or is there more to it?
01:26:58
Speaker
Yeah. I mean, at the end of the day, we have a very clear idea in mind for what these machines have to be able to do in order to compete with fracking. Right. And so do they need to last 50 for 50 years to compete with fracking? No. Right. Do they have to look beautiful to compete with fracking? No.
01:27:16
Speaker
What do they have to do? They have to produce natural gas as cheaply as possible. They have to pay for themselves as quickly as possible. That's it. That's the only things they have to do. And everything else is subsidiary or subservient to those goals. And if you internalize that enough, it drives you decision-making and it drives it in the correct direction. Yeah.
Reflections and Inspirations
01:27:40
Speaker
I realize we're coming up to probably an hour and a half talking and while you told me you don't have a hard stop, I feel like I will bounce pretty soon, unfortunately.
01:27:51
Speaker
Yeah, I'm certainly limiting the amount of time that you're spending making these improvements or building these machines. And therefore, there's probably some carbon costs to every word that I'm saying here. No, I think it's all upside. It's all upside because if one of your listeners is inspired to go off and try something, it will have been worth that time. Right? That's all it takes.
01:28:18
Speaker
Well, I can think of more things to discuss, but I don't think I can think of a better point to end on. So Casey Hammer, thank you so much for this conversation. I know you have a lot of problems to solve and I think you're going to do it. Well, I appreciate that. Thanks so much for having me on your show. It's been a pleasure.
01:28:50
Speaker
So,