Hacker News Reader

Comment by immmmmm | Hacker News Reader

I wish that the 3 law of thermodynamics laws were more commonly taught... because here you are battling the 2nd and the 1st.

A calculation by Jean Marc Jancovici showed that for a small airport (GVA in this case) you'd need half a dozen nuclear reactors just to produce the fuel for departing flights. That's obviously assuming this kind of technology is functioning and has 100% yield.

I'm not saying that these technologies are not to be pursued, but thinking that we (the rich) can keep flying as much as now due to a miracle technology is unsupported by Science, to say the least.

ps: i did 2 postdocs in material science trying to improve various industrial/energy technologies. there's no miracle.

View on HN

I don’t think the goal is to keep flying as much, but to be able to fly at all.

Also electrifying entire trucking and car fleets will take decades.

This provides a means of producing an incredibly densely stored source of energy from abundant inputs, that is the same price as oil.

It can solve for the problem of net zero production and transportation of equipment used in renewable energy production, for example, using present day transportation technology.

There are no silver bullets, but all of these innovations add together in order to create a combined solution to the worlds energy problems.

> There are no silver bullets, but all of these innovations add together in order to create a combined solution to the worlds energy problems.

i worked all my professional carrer in science and innovation, if you have a reference for that statement i'd be glad you share it. maybe i missed something.

Do you mean that there is a silver bullet,or that innovations don't add together to provide solutions?

there's none. and (unfortunately for me) the best solutions often involve less technology rather than more.

If fuel from air can be made efficient, then combined with highly economical fusion, I could see it being viable. I'm not sure it will ever be cost competitive with long tail oil and gas, though. And that's at least 30 years out.

> then combined with highly economical fusion

If we had some ham, we could make ham and eggs, if we had some eggs.

"Highly economical fusion" is probably an oxymoron.

There are multiple companies closing in on highly economical fusion. Enough so that they're receiving hundreds of millions of dollars of investment from very savy private investors. My money is on Helion getting there first.

Are you aware that tritium like hydrogen is very hard to contain as it's as small as hydrogen and very nasty as it's highly radioactive. That alone prevents this technology to become cheap any time soon (assuming they get Q>1).

None of the DT approaches are going to be highly economical, for fundamental reasons. I do agree that Helion is the least dubious contender.

Solar and wind will be effectively free in the next decade compared to any other energy source with realistic projections as low as $10/MWh in high production areas and lots of promising battery technologies that may be able to timeshift energy for a similar cost by hours or days.

As we reign in the gratuitous subsidies to the fossil fuel sector, the question then becomes can we provide the plant and operational costs to store electricity chemically for $30/MWh (or as fossil fuels get more difficult to extract, $60 or $100).

Given that it's a brand new industry and you can do it right now with hydrogen for $200, energy included, then we only need to price in a fraction of the externalities right now and airline operators are going to be looking pretty hard at various green fuel options.

I think the goal should be to be able to fly as much and more.

Why is commercial aviation as a form of transport a useful goal? There are alternatives that have realistic energy efficiencies and use available technologies, like rail and nuclear shipping.

yes! rail is amazing in term of efficiency, a common high speed train has around a credit card worth of contact area with the rail for the whole train! it's also quite relaxing and socially interesting.

the problem is that rail is "boring" and rarely get sufficient funding.

It also allows for much more productivity while travelling, which is important for business travellers.

Rail is quite a lot less flexible than air travel. If you have a plane you can pretty much go anywhere that has a strip of asfalt of the right length. With rail you need to lay out the track between the two points with all the difficulties involved in that.

It's wired vs wireless ethernet. Which do you have in your home?

Energy efficiency will become radically less important as energy cost continues on down.

Then other things become more important, and drive choices.

There are no useful alternatives to airplanes for crossing a continent, or an ocean.

I agree we should do more trains, but they are limited to around 1500km per trip before flying is enough better that you look like a fool for suggesting it.

It would be slower, yes, but you could still get from London to Beijing in a few days, which seems reasonable when you look at historical travel patterns and the sheer inefficiency of the alternatives.

This is not 1290 when Marco Polo was traveling around. This is 2022 when people have seen much faster and nicer alternatives. Nobody is going to put up with days of travel when they can fly anymore.

I'm not convinced trains are more efficient than planes at that distance either - I don't' know how to analyses it, but planes see much less wind resistance because of altitude.

In more traditional units:

Prometheus can turn 77kwh of electrcity into 1 gallon of fuel.

Jet plane with range of about 7000km has 52000 gallons so roughly

4GWh of electrical energy in to make the fuel, about half of which is lost as heat and the other half moves the plane.

So the real question is, do you have a better use for that 4GWh of electrical energy? (Or more generally, and back to car size would you prefer 77Kwh of electricity or a gallon of fuel?)

Carbon free electricity so cheap that you can make carbon containing fuels for less than fossil fuels is like a very heavy carbon tax in impact. Yes you can do whatever you like with the fuels, but since you pay directly for it, the market will respond by finding cheaper and more efficient ways to do the same thing, using the electricity directly or as a cheaper to convert fuel.

Their pitch kind of makes sense for long distance flight, but unless it's a classic car that you only drive once a year, this fuel will have the exact same impact as a heavy carbon tax on the market for motor vehicles i.e. they'll be scrapped and replaced with EVs that are cheaper and cleaner to run.

Solar is cheaper than nuclear and still getting cheaper, so looking at what you'd need in terms of nuclear reactors is likely to be misleading.

I don't even want to think about how much surface area you'd need for solar panels intended to replace 6 megawatt-scale nuclear reactors, but I'm guessing football fields is the wrong unit to use...

Last time I checked, land wasn't cheap and they weren't making any more of it.

yeah and there's crop failure coming decades in advance compared to IPCC's predictions, so better not using crop space.

Some of that crop failure is due to heat, which could be mitigated by using solar panels to provide shade. That doesn't work for all crops, some of which require direct sun, but it works for more than I expected.

No crops require 100% direct sun.

Maize and wheat yield is reduced a bit, without, but the year-round revenue from solar panels easily exceeds the difference.

Land isn't cheap, but there's a lot of land that can have a solar panel on top of it without causing problems. A good place to start is every building roof.

Land is, in fact, cheap in very many places. If one were making fuel with the solar energy you wouldn't put the solar in places where land was expensive, since the fuel would be highly transportable.

> I don't even want to think about...

I'd call that an argument from laziness.

> Last time I checked, land wasn't cheap and they weren't making any more of it.

Land that is remote, not fertile and that doesn't harbor any natural resources is actually rather cheap. Moreover, "they" are making more land in certain coastal areas where land is expensive.

Regarding land use, just yesterday I was pondering about what technology it would take to create survivable off shore solar. Some loose grid of floating collectors happily bouncing on the waves like a flock of resting seabirds, perhaps cleverly reeling in and out link and anchor lines to match the geometry of the waves? Or just the right amount of springyness, dynamically tuned to the wave situation?

Then it occurred to me that even nature hasn't really solved ocean surface plants, what could be a more clear indicator that it's a really hard problem...

We have plenty of calm freshwater sites for solar, so there will be no need for ocean solar.

The correct unit to use is parking lots. If you base this unit on the current area used for parking in the USA, you can even use centi-parkinglots ;-)

SF Reservoir has 12 football fields worth of solar panels and generates 4.5 megawatts.

It generates zero megawatts far more often than 4.5 megawatts. 4.5 is peak on the best days. 0 is assured for several hours every day.

Thanks for added details. I was wondering if this project was efficient considering the amount of "fog" in this area.

When algee biofuels were first proposed someone calculated that Arizona could produce all the fuel we use on currently unused land.

The limit is our ability to scale algee, but the lab results also show that the numbers are right.

Yes, you didn't think about how much surface area, since if you did you'd realize it wasn't a big problem.

(I assume you mean GW not MW, because MW is very little for a proper power plant). Solar panels produce about 200Wp/sqm. In Germany they average to about 12% of peak production over a year, so say 22W/sqm averaged. 6GW continuous is then about 51k football fields[1]. I assume that the US has lots of land that is much sunnier than Germany, so you can probably get away with fewer football fields.

[1] https://www.wolframalpha.com/input?i=%286GW%2F%2820W+per+squ...

What area would it take to produce several GW that a few reactors would produce?

In clear weather, a surface perpendicular to the sunlight gets about 1 kW / m². So 1 GW would take a square kilometer. If our solar cells are top-notch 25% efficient cells, 1 GW will take 4 km² at best weather. And we need many multiples of that.

Verily, Sahara must be the best place for such an oil factory. Maybe some of the US Midwest and West, with many sunny days. And, unlike the cute pictures on the side, they will need to be massive, all the way to the horizon.

Maybe with a few nuclear plants thrown in if we learn to build them efficiently again.

> What area would it take to produce several GW that a few reactors would produce?

Out of curiosity, I tried some napkin math on this. If Nevada went whole hog, used say, 1% of their total land for solar, they could produce almost half the electrical needs of the US, if I haven't mucked up the math too much.

Sounds plausible, but 1% of Nevada's land for solar would indeed be 'all the way to the horizon' in many places.

That number is a bit low, but area for solar panels will never be in short supply, because the same land can be used as is used for other things.

let's assume we can scale renewables+nuclear fast enough so that our civilisation is not damaged too much by climate change.

let's also assume we stabilised biodiversity and avoided a mass extinction.

should we continue to produce massive amounts of energy so that some can commute by plane daily and others can stream cute kitten video in 8k?

If all that were true, I'd say definitely, we earned it. Make it a 80k kitten!

If we can continue producing them in a renewable or at least sustainable way, why not?

If the only way to keep that up is burning oil, then no.

Please also note that some of the most intense producers of CO2 are poor(er) countries which burn coal because it's cheap. Even natural gas can be too expensive for them, let alone solar or wind installations.

I wonder when the West would consider buying and converting such plants. It's likely feasible in Africa, hardly so in China.

I looked at a 2GW nuclear facility, and it's about 1 square Km. The reactors aren't too big, but there a lot of auxiliary buildings as well as parking for many employees. I know one building is a training facility that replicates much of the primary facilities, and there is a lot of space dedicated to security either with actual structures or buffer zones. There are also PR/education facilities on site, presumably so the local municipality doesn't eventually vote them out. Somewhere they store fissile material.

It's also on the coast for access to coolant water.

It seems very difficult to place huge facilities like this with all the staff and security requirements on waterfront property when you can stick a solar panel on or next to just about anything.

If you reframe it with solar panels it is just as troubling. Because then you need a massive amount of area for the panels.

Or as their competitor puts it, we're going to need a lot of solar panels.

https://terraformindustries.wordpress.com/2022/07/24/were-go...

a few dozen of km2 according to my napkin calculation. for one minuscule airport.

The proliferation of solar farms is well underway and not abating with or without e-fuels, this just gives greater optionality to the energy they output and helps with intermittency. The land-use footprint of their tech as such is 100,000 gallons of car fuel per year per 'forge', each of which fits on a flatbed truck, plus a slated 5 manufacturing facilities to make the forges. That doesn't seem too bad.

On the land-use of the solar panels themselves, I wonder if at some point in the future it might be possible to beam a directed laser to earth from space-based solar arrays, so the earth-based footprint is reduced.

Show your math. Unless those nukes are very small, your claim seems way off.

My back of the envelope calculation is that 1GW for 1 year is 30PJ. Jet fuel has 42MJ/kg with a density of 0.8 kg/L for a total of around 100 million liters of fuel at 100% efficiency. An A321 holds around 30000L which comes out to about 30000 flights equivalent from one reactor. GVA had around 200000 flights in 2018 meaning about 6 1GW reactors equivalent of fuel used (obviously not all flights would be fully loaded but I don't know what a normal load is).

Another quick calculation, in addition to the one from tfussell:

Google tells me [0] that total fuel consumption by commercial airlines in 2019 was 95 billion gallons. One gallon of fuel has around 33 kWh of energy.

Power capacity needed to produce that amount of energy in a year is around 358 GW.

Total world electricity production in 2020 was around 3000 GW-year [1].

[0] https://www.google.com/search?q=total+yearly+aircraft+fuel+u...

[1] https://www.google.com/search?q=total+world+electricity+prod...

Comparing global fuel consumption with global electricity production is a good approach. It’s clearly substantial, but doable, especially since fuel production can utilize “unreliable” renewables (make gas when the sun shines).

According to this source [1] flights departing GVA generated 1.3 million tons of CO2 in 2018. According to this source [2], jet fuel generates 3.16 kg CO2 per kg of fuel. Multiplied by approximate fuel density and we get 2.57 kg CO2 per liter. We can therefore estimate that GVA consumes approximately 500 million (1.3 billion kg CO2/ 2.57 kg CO2 per liter) liters of fuel per year. Jet fuel contains about 35 MJ of energy per liter according to [3] so that's about 17.5 PJ. If the process to convert electricity to jet fuel is 50% efficient, that's 35 PJ. That is equivalent to a 1.1GW reactor running at 100% capacity. At the global average capacity factor of 80%, that's about 1.4GW required. Half a dozen is probably a pretty large overestimate. Alternatively, this would require around 6GW of solar, although 6GW of solar is probably quite a bit cheaper than 1.1 GW of nuclear power.

I forgot to add the sources earlier. Here they are now: [1]: https://rapports.gva.ch/rdd/2019/en/Article/minimiser-l-empr... [2]: https://www.eesi.org/papers/view/fact-sheet-the-growth-in-gr.... [3]: https://en.m.wikipedia.org/wiki/Energy_density

agreed, 1.1 GW seems reasonable for GVA. thanks!

I agree it was overestimated, note to self: always redo ppl's maths.

1) number of barrels of jet fuel per day in switzerland:

34000 barrel / day = 0.39 barrel / s

https://www.indexmundi.com/energy/?product=jet-fuel&graph=co...

2) energy in a barrel of jet fuel:

1700 kWh / barrel = 6120000000 J / barrel

3) total power for switzerland:

2.4 GW

so more like 2-4 reactors for all switzerland.

Reply on news.ycombinator.com