Launch HN: Prometheus (YC W19) – Remove CO2 from Air and Turn It into Gasoline
An article about us came up on HN recently and people seemed interested (https://news.ycombinator.com/item?id=19792412), so we thought it would be good to try to answer some of the questions we saw there and try to dive in some more to any questions that follow!
The only inputs to make the fuel are CO2 and water (both from the air) and electricity. The only outputs are fuel and oxygen. One way to think about it is that making fuel is reverse combustion. The process isn’t super efficient (we expect 50-60% overall efficiency at maturity), but it turns out that doesn’t matter as long as the electricity is zero carbon and low cost. If the cost of our equipment is also low, then we believe we can not only make zero carbon fuel, but actually compete on price with fossil fuel.
We’re not the first to make fuel from the air - in fact Google, Audi, Carbon Engineering, Global Thermostat, Climeworks, and labs at universities and national labs have all done it before us. What no one has been able to do so far is do it at a low enough cost to compete with fossil fuel.
The thing that’s new about what we’re doing is that we have gotten rid of all the thermal processes normally used, and instead use a process that uses only electricity (no natural gas, etc) and does it at room temperature. This is a big deal for both capital cost and for being truly carbon zero. We can use inexpensive materials, which keeps our cost low, and can start up and shut down quickly, which allows us to run intermittently, matching the intermittent nature of many renewable energy sources. We can also only run when the power is at the price we want.
Digging in to some more details, we absorb CO2 and water vapor from the air into an aqueous electrolyte. We then react the CO2 in the water with a copper catalyst to directly make alcohols like ethanol, butanol, propanol, etc. Both of these things have been done by many others and the science is known. Normally at this point one would have to use a thermal process (distillation) to get the fuel out of the water, and this is expensive and makes the economics really hard to get right. We don’t have to do this step thermally though, because we have a carbon nanotube membrane that replaces it, extracting the alcohols from water in a single step at room temperature. This makes a huge difference in cost. The last step is that we up-convert the alcohols to gasoline, diesel, and jet fuel. This last step is also well known and we can actually buy this step from others.
The carbon nanotube membrane that makes this all work is the product of 6 years at my previous startup, Mattershift. I was developing it for desalination and water purification. About 3 years ago I realized it could do this job, but it wasn’t clear that a startup could raise money for such an ambitious effort, especially one linked to a political issue (unfortunately) like climate change. When I saw the YC request for startups in carbon removal, I knew that the timing was right, and I founded Prometheus to do it.
Please let me know if you have more questions or feedback. I’ll do my best to answer any questions, but please excuse if I’m not able to go too far into details like our piping and instrumentation design, or other really specific things we wouldn’t want to help competitors with.
Thanks!
588 comments
[ 2.5 ms ] story [ 278 ms ] threadJust guessing that real estate is going to be a consideration for this venture. Where (geographically) do you plan on operating? Is there any advantage to doing this in areas like Los Angeles/ Mexico City that naturally trap automobile emissions?
Or even having multiple points of capture across a geographical area.
Put that way, it's a huge improvement and avoids having to electrify entire fleets first.
Another thing, will this synthetic diesel burn cleanly like hydrogen or do they need to add lots of nasty stuff?
Obviously there are lots of benefits of batteries over gas, but being able to load up a standard 10,000 gallon tanker truck with 300MWh of energy and drive it anywhere there's a road is something batteries simply can't do.
Maybe I'd encourage you to think of yourself less as a fuel company and more as a chemical precursor one. Outside of combustion, there are a ton of other use cases in need of synthetic pathways.
Are there any other products remaining after the conversion from alcohols such as propane, butane etc..?
- I'm not a chemist, maybe you said this implicitly, but roughly at what rate would this capture a kilogram of carbon?
- Why did you choose the name 'Prometheus' the titan who created man, gave us fire, and as punishment for the latter part, had his liver eaten out by a vulture every day for eternity?
Has?
Second question - I chose Prometheus because the story is that he gave us fire, and we have a responsibility to use it well. So far, not so good, but now we can do better. I imagine him saying "here's fire, don't screw it up".
What's the expected total removal per year in, say, your first, fifth, or tenth years of operation?
It does reduce or eliminate the need to source gasoline from "fossil" sources however.
Maybe a non-membrane version could be attempted that used renewable energy (sunlight) to do the boiling? Certain geographic regions might have the right combination of plentiful sunlight, water, and wind to minimize all these costs?
You said you're using the nanotube membrane to separate petro products from the water. What do you do with the water left over? What state is the water in when the process is complete? Potable? In need of treatment by traditional water treatment facilities?
This implies there is still water there. Maybe it was poor phrasing, and all of the water is consumed, but that isn't what is stated.
This implies there is still water there. Maybe it was poor phrasing, and all of the water is consumed, but that isn't what is stated.
Assuming some water is lost due to electrolysis, you can top it off over time.
If the catalyst breaks down, or some undesired reaction limits efficiency (i.e. generation of perioxide or other undesired chemicals that degrade or react with the membrane, electrodes, catalyst, etc) then I think there might be some treatment or maintenance necessary.
However, none of this sounds remotely as bad as the treatment needed in the production of petrochemicals or lithium batteries or recycling.
Your question is, in the distillation step, the nanotube membrane separates fuel from water, leaving some amount of water; what happens to that water in the process?
The question that was answered was, what happens to waste water? For which the answer is that there is no water "left over", and so the question of what happens to it is ill-defined because it doesn't exist.
So I guess what is happening is that any water left in the aqueous electrolyte can continue to be used in the process indefinitely, and there is no point at which the water becomes waste water? That all the water collected from the air gets broken up into hydrogen (in the fuel) and oxygen (waste product, sent into the atmosphere) and the amount of water in the aqueous electrolyte doesn't grow?
In the near future, cars and trucks will be electric. This is shuffling deckchairs on the titanic. Why not put 100% of your efforts into electric vehicles powered by 100% renewable sources?
All fixable things, but will take tens of years. Let me use carbon neutral petrol
The idea and processes behind Prometheus are still very interesting and have implications beyond the short-term focus on carbon footprint. This type of process will likely be needed to manufacture fuel on other planets.
I agree we need to do something now, but this is just putting off the inevitable. Re-engineering dirty fuel sources doesn't buy us time to resolve this. It is procrastinating, shuffling titanic deckchairs.
The main advantage of this tech I can see is making 'primary' sourced fuel illegal, 'secondary' sourced fuel like this prohibitively expensive, and forcing everyone to go pure electric engine by economy.
Such technology would make gasoline have no output, since the ouptuts would be reabsorbed. If anything, electric cars, due to the massive environmental damage due to rare earth mineral mining, would be more of an environmental disaster.
It sounds like you're suggesting that burning the fuel releases more carbon than was used to produce it, which isn't possible.
I don't understand your rail/car replacing flying argument. There are no signs that flying is going to get so expensive that people would rather pay for it with so much of their time.
Using a process like this would essentially make the fuel simply a much better battery for airplanes than our batteries today.
Electric cars can't compete against gasoline cars until the price hits about $6/gallon. And considering that crude prices are relatively low, but gas in CA is almost $4/gallon in the Bay Area, it doesn't take much time for normal rates to hit $6/gallon in the next 10 years or so.
At that price, it becomes more economic and that's the time frame probably what they're targeting as well.
And the national average (today) is $2.95, you might as well compare Hawaii and Alaska which are also economic islands.
>it doesn't take much time for normal rates to hit $6/gallon in the next 10 years or so.
The national average 10 years ago was $2.45, which is $2.90 accounting for inflation. The only way gas hits $6 is if we have a second Great Inflation or massive political changes in the United States and abroad.
By that measure, Florida or Oregon would be economic islands due to their anomalous taxes.
no, we're taking Soothing30MinuteRelaxingBreaksAt200MileIntervalsAwesomeHappy™ brought to you by Tesla™ instead
It's there when I need it for the long road trip. But daily life is effectively refueling-free.
What percentage of round-trips are under 200 miles? NHTS recorded the distance from ~750,000 car trips in 2009 and found the 95th percentile is 30 miles, and the 99th percentile is 70 miles. Somewhere on the order of 1 in 1,000 trips needs a Supercharger stop along the way.
[1] - https://nhts.ornl.gov/vehicle-trips
IF this is as good as they claim I don't see electric cars as worth it as a first car. They are slightly cheaper for an in-town only commuter car, but the downsides of electric cars (expensive battery replacement, limited range without long recharge times) are something that fans hate to acknowledge. Note that I said first car, once you have decided to have a second car electric is probably better.
With V3 Supercharging, you can drop in with your Model 3 at 15% and be at 80% in 24 minutes. In my road trip experience, that's barely enough time for everyone to use the bathroom and pick out a beverage. [2]
[1] - https://www.tesloop.com/blog/2018/7/16/tesloops-tesla-model-...
[2] - https://twitter.com/privater/status/1103567772301193216/phot...
So if you are starting the day with 95% charge, and ending the day plugged in at your destination with 15% charge, that means you can travel about 450 miles with just one ~20 minute stop to charge along the way.
How is that not fast enough? The only issue is if you can't plug in at your destination. Maybe in that case you have breakfast near the next Supercharge on your route.
Yes, it is something you might have to think about for a minute. The guidance software will map out Supercharging stops for you.
I do expect that they will figure out a reservation system which tracks and queues vehicles en route, like an air traffic controller, to cut down on inefficient ad hoc queueing at the location.
It would be super fun to program such a system, backend and front. Even more so as the cars become increasingly autonomous.
For now at least you can click on the map and see live how many chargers are in use.
You can wait around for 12h to charge, or you can turn in your cores and pay for 100% capacity batts now. And the swap would be quicker than standing around filling up a tank.
But no. Each ev is different and proprietary. The chargers aren't even the same.
https://www.greencarreports.com/news/1090933_standardized-el...
So swapping batteries is hard to do, and unnecessary for most drives. Sure, you’d want to be able to swap batteries in five minutes like pumping gas. But who’s going to ship batteries out to the middle of Kansas, build infrastructure, and pay technicians just for people on a road trip who want to swap batteries?
But you’d still need to have expensive batteries in stock on location, technicians and equipment. I can’t see that ever being done at scale.
For day to day use EV doesn't need trips for gas station to refuel - so you actually save time.
For long trips - yes.
Because I keep track of my mileage and have to carefully babysit the process (likes to overflow) I’d say I lose at least 10 minutes every time I get gas, and that’s every 130 miles or so.
Only 130 miles? Is that normal? My car can easily go 600 miles on a 15 gallon tank (diesel), but I've had it for so long I don't remember what other cars are like.
Now I'm curious how long it actually takes. Next time I fill up I'll time it.
The DOE's 2020 targets came three years ahead of schedule. Their 2030 target is $0.04/kWh [2], which would work out to $2.40/gallon. Mix in the fact that a bunch of countries (and US states) have implemented carbon taxes and you've got yourself a good long term investment.
[1] http://solarcellcentral.com/cost_page.html [2] https://www.energy.gov/eere/solar/articles/2020-utility-scal...
Taxes are low as here in the US (a) many people hate taxes and (b) with such a low population density, commerce is very reliant on road vehicles. So raising taxes has an outsized impact on commerce.
It's just a great storage mechanism.
Storage is also a big issue when it comes to renewables. You could take in carbon in areas with lots of sun and ship the fuel to areas not well suited for solar generation.
This really would be a game changer for lowing the net carbon output.
This also means you could solve one of the big problems in the power grid, peak generation. Use the extra capacity during off peak hours to generate fuel that is later used to fire up power stations to supply peak demand.
A gallon of petrol - 40 miles (realistcally and using UK gallon).
According to Tesla their charging is 92% efficient so reduce that 200 to 184 miles.
Your process is 4.5 times less efficient than just putting that electricity into the EV? Is that right?
If so - and your process is carbon neutral (big if) - what's the point in a future where EVs dominate?
EVs won't dominate some important uses for a long time, e.g. aviation and marine shipping.
It might also be worth looking at where airplanes tanker fuel, that is to say, carry more fuel than they need for their current leg because refueling at the next stop would be difficult or expensive. Apparently a lot of that currently happens on short flights to small islands; while I hope Wright Electric and/or the EViation Alice will eventually take over that market, in the short term that's a market that might be willing to pay a bit more for liquid fuel made from air plus local solar panels.
Fuel oil that can be burned in combined cycle power plants in the winter may also be valuable for dealing with seasonal imbalances in demand vs renewable generation that lithium ion batteries can't cost effectively balance.
Assuming they get to 50% efficiency, that's $1.69/gallon.
Sounds crazy low, but I hope it's true.
[1] https://www.utilitydive.com/news/texas-muni-signs-cheap-sola...
My view is that is more a question of when than if and that the point where it stops mattering relative to the comparison to fossil fuels depending on your point may already be quite near or even in the past.
Some places are still getting expensive solar, some places are bidding at 0.025/kwh. It doesn't really matter. What matters is what the bids will be ten years or so from now, which is when realistically this could start becoming operational at a meaningful scale.
It's 2019. Ramping up production for something like this takes some time. A decade can fly by for a startup like this. Doing the math with 2019 prices and efficiencies means you get a very conservative view of what would be possible now with today's level of technology at today's scale.
However, an investor needs to look a decade ahead. Or longer and assign some probabilities to likely outcomes. The outcome where there's no progress whatsoever in making clean energy tech better in the next ten years seems unlikely. So, betting on < 1$/gallon as a feasible goal in 1-2 decades seems like a reasonable bet. If the rest of the technology works as advertised (room temperature synthesis and extraction of alcohols) and the technology can be scaled at reasonable cost, that ought to be basically a money printing machine. The lower that price drops, the better. As long as oil remains the primary source, you can pocket the difference as profit.
In the UK, people pay approx $6 per gallon for gasoline.
Of that, AUD 0.41 is federal fuel excise, and AUD 0.12 is GST.
If you change that chart to USD and Gallons you'll find this around the cost of gasoline in New Zealand.
China’s EV goal is 20% of new vehicle sales by 2025. That is to say, 80% of sales in 2025 will still be gas. (So will the 100 million gas vehicles sold in China between now and then). We are a long ways away from not needing this tech.
A lot of people in the world. And Americans would pay close to that without subsidies + rising gas prices.
Today's price for standard gasoline here in the Netherlands is 1.83 EUR/litre, which is currently equivalent to 7.75 USD/gallon.
[1] http://www.hydroquebec.com/business/customer-space/rates/rat...
Specifically; How many barrels of ready to burn gasoline are you outputting per day in 6 months? 1 year? 3 years?
The fact that the general fuel is going to be sold and re-burned means this is at best carbon neutral, so from a cynical environmentalist point of view, what's the point? This can't help anyone unless you pump the synth-gas into the ground and sequester it, but that isn't VC-backed viable.
If you are using some sort of membrane to pull out distillates without heat, why wouldn't you just sell that to existing petro-chemical companies for their distillation processes? What non-thermal-energy related benefits come with this membrane?
The poster gave important information in another thread: 60kWh per gallon of gas, aiming for ~5 cents per kWh, ~$3.00 per gallon into distribution channels. That makes it instantly cheaper than petrol in, at minimum, the UK
The status quo today is that we are removing carbon from the ground and putting it into the air. This replaces those fuels that currently come out of the ground. It's still a big win, even if it's technically only carbon neutral.
Even electric vehicles aren't carbon neutral, because they're mostly charged by coal power.
Also: you suggest the synth-gas should be pumped into the ground. What would be the point of converting the carbon dioxide to alcohol to gasoline and spending electricity to do it? At a minimum you could just pump at the alcohol stage.
> Even electric vehicles aren't carbon neutral, because they're mostly charged by coal power.Even electric vehicles aren't carbon neutral, because they're mostly charged by coal power.
Is thus an invalid/irrelevant argument.
I can get excited about jet fuel however, since battery powered planes are considerably less practical than their traditional counter parts with the current technology.
I assume the company is operating in the US. In the US only 27% of electricity is from coal and falling. https://www.eia.gov/electricity/monthly/epm_table_grapher.ph...
And in California (where nearly half of the US's electric vehicles reside--with EVs achieving a nearly 8% market share last year among light vehicles), it's MUCH less than that. There's only a single, tiny 63MW coal co-generation plant in the state, so coal power is less than 1% of California electricity.
> Even electric vehicles aren't carbon neutral, because they're mostly charged by coal power.
Additionally, it's absurd to compare two consumers of electricity, both of which are flexible with respect to time of consumption, and just arbitrarily say one is mostly coal power (which is false) and the other isn't.
We definitely need carbon neutral or even carbon negative industrial processes. I kind of question the logic of going for fuels versus more valuable non-fuel feedstocks, however. A maybe 50% efficient conversion of electricity to alcohols with a typical 20% energy conversion of fuels to mechanical power gives you about an order of magnitude less efficiency than battery-electric.
And if you ARE insistent on making carbon neutral fuels: Makes more sense to focus on things that are super hard to electrify, like long-haul aircraft and transoceanic shipping and maybe heating fuels for the far north (where air source heat pumps lose their effectiveness in January).
But I'm glad an electrochemical process is being developed, here. Electrolysis and related processes are much preferable, IMHO, than the more common thermal processes.
Sounds like a great idea! My general point is that zero carbon energy sources like wind and solar are at least mildly zero sum, and creating brand new load on them would likely just result in capacity moving over to other producers like natural gas and coal.
I guess I don't understand why YC is throwing money at this because I'm not confident there's profitability on the horizon, unless they nail down mind bogglingly large scale of production of aviation fuel,
Meanwhile, if San Fran is looking to dump money on (IMO) nebulous and green projects, give me money and I'll plant 100k trees
If they can scale their process quickly (not just on-off, but to levels in between) the power company will be happy to give them a substantial discount. They can probably get power for $.03/kwh - windmills are about $.02 for profit so that is plenty of profit to pay for transmission infrastructure. Of course they have to agree to how quickly they can change their production.
This assumes they can turn off completely at 5pm when everyone gets home and jumps in the shower needing hot water. Then go to full production as they go to bed while the wind blows the best. Then drop down overnight. Then scaling up again as the sun comes up. In the late afternoon as the sun is hot they scale down to account for air conditioning. Of course they may as well shut down for maintenance in December (Christmas lights). If they can pull this off renewable energy can scale grow to be much better while taking more power plants offline. That is a big if though.
What steps are in place to prevent the process from displacing other renewable use? Take the fact that it is less than 100% efficient to store renewable energy as gasoline. Others might make more efficient use of this renewable energy by immediately using it to displace fossil fuels, with no or lower storage losses. Under this scenario Prometheus is again a net carbon generator. (edit: The same dilemma is faced by any other type of storage.)
It strikes me that Prometheus will come into its own if it can guarantee that it doesn't displace more efficient users. This will happen when fossil fuel usage becomes negligible or we (as a society) come up with a sophisticated energy scheduling algorithm that can take account of efficiency.
It comes back to that fact that whilst it is green to replace fossil with renewable energy, it's even greener if we can increase efficiency and avoid consuming that energy in the first place.
Don't get me wrong, correctly used it's a great thing you are doing.
A couple of scenarios where this is ideal:
1. Located close to intermittent renewable generation that is already hooked up to the grid, especially in places like California where intermittent renewables can already push generation beyond load at peak times and result in near zero or negative prices, this can soak up electricity usefully in a way that displaces other carbon emitting fuel.
2. Deployed along with intermittent renewables in the most optimal places for generation, where connection to the grid is impossible, inefficient, or will simply take time. This can soak up the most efficient possible generation of deployable renewables which would make otherwise uneconomical utility scale deployment viable.
Both of these are scenarios that our capitalist market system should work well to optimize for assuming the electricity->fuel technology reaches suitable efficiency (and not before then).
Also, fuel is fairly transportable. You can produce the fuel wherever renewable electricity is most available, and sell it worldwide. The economics are comparable to aluminum, where a large fraction of the world's supply is produced adjacent to geothermal power stations. https://en.wikipedia.org/wiki/Economy_of_Iceland#Aluminium
It's meant to replace a carbon-emitting process (burning gasoline made from fossil fuels) with a carbon-neutral process. The replacement process is not carbon-neutral.
The membrane will definitely replace distillation in a number of industries, but the most exiting thing to do with it is to replace fossil transportation fuel with renewable fuel.
Nitpick: It will leave carbon emissions unchanged, but may reduce the amount added to the carbon cycle.
What kind of production quantity are you looking at? Small batch aviation fuel for a niche "green" airline? Every semi truck on the road in America? The neighbor kid's go-kart?
You give an "efficiency" of 60ish %, but I don't know how to interpret that. How many megajoules (or kilowatts) are required to produce a gallon of 86 rated gasoline?
Think stuff like big jets, or container ships. If you sit down and crunch the numbers for any non-carbon energy carrier, you either end up sinking the ship or with a fuel tank twice the size of the cargo.
[1] https://news.ycombinator.com/item?id=19845427
But I agree container ships would be a good application of this. Container ships also have pretty high efficiency internal combustion engines (about 50%), meaning there's less advantage in electrification there than for land vehicles which often have 15-25% efficient engines. And it allows low particulates, which is one of the main forcing regulations on shipping powertrains nowadays, without the overhead of cryogenic LNG.
However, container ships use ridiculously cheap bunker fuel instead of highly refined gasoline.
The effect (not counting contrails) is about double the effect of CO2 emissions alone. Including contrails, it may be even higher. There are mitigation strategies to deal with these non-CO2 radiative forcing effects, though it's very complicated.
Good article: https://www.carbonbrief.org/explainer-challenge-tackling-avi...
https://elib.dlr.de/68051/1/fugl-2010-4648.pdf
which says
> At pressures greater than 500 hPa (i.e. below roughly 6 km), the forcing [of water vapor] is assumed to be negligible
So flying below 6km would help a lot for impact of water vapor. However, given this was not yet acted upon, IPCC is probably right that
> the effect of water vapour emissions is likely to be a significant, or even the dominant, contributor to their climate forcing.
This is even worse given the projections of aviation expansion. Right now the water does not seem to be a big problem and it could stay that way - water lifetimes even in stratosphere is counted in years, as opposed to centuries for CO2 - but if aviation expands, then it becomes more serious than CO2.
This is an utterly asinine question. The point is that most current fuel is not carbon neutral so this is an improvement over the status quo.
This doesn't make sense to me? Wouldn't even the most cynical environmentalist agree that carbon neutral is much preferable to carbon positive?
This thread is such a nice break from that, feels so refreshing. I don't know anything about this process but I hope it works out.
That's a lot of "if". How close are you to that?
If you plan to sell to car users, how do you plan to compete with electric cars, that can get about 8x the use out of the electricity (and don't have most of the many issues of combustion engines).
Am I missing something really obvious? Is transport through cable for direct consumption prohibitively expensive from the energy sources you're planning to use? Do fuel cells change the calculation? What's the advantage? Potentially take over a rapidly dying market?
Large sea shipping, aircraft, and rockets will continue to use fuel for a long time.
Please keep us all updated on how it goes, I'm sure you will see a lot of excitement about what you are doing.
I'm excited to see what comes of this in the coming years!
My thinking here is that creating gasoline and selling it for immediate use is at best carbon neutral, and at worst delaying the replacement of that gasoline-burning equipment with an all-electric version. But if your focus was on replenishing raw oil reserves for future (hopefully mostly non-burning) use, you'll be removing CO2 from the atmosphere. To do it at significant scale requires a lot of energy, and a nuclear reactor is much better at providing that level of energy than other zero-carbon sources. The trade-off is nuclear waste, but that's much more containable than the waste from non-zero-carbon energy producers for equivalent amounts of energy. Eventually it could be reprocessed to extract more energy from it as well, until the radioactivity is used up.
What efficiency are you at now and what happens to your startup if you never reach that "maybe too optimistic" 50-60% goal?