Hacker News Reader

Comment by xoa | Hacker News Reader

I'll admit I really don't get the fascination with hydrogen as an energy storage medium. It has to be manufactured, so how green it is depends on whatever the power input is coming from. It's not any inherently greener then synthetic hydrocarbon production, which can be produced from atmospheric carbon for zero net carbon impact, and of course has no sulfur or aromatics. It also works fine with existing engineering. And hydrogen is an absolute total bitch to work with. It's got bad density, a really low/really high pressure liquidation point, escapes from everything, embrittles steel, and on and on. The infrastructure has no multiple use, unlike improvements to the grid.

So why not just do full manufacture of regular fuel? I've never understood where hydrogen fits between improving batteries/capacitors and synthetic hydrocarbons. It always seemed like an answer without a question outside of rockets (and even there the economics and practicalities of methalox may be better).

View on HN

> It's not any inherently greener then synthetic hydrocarbon production, which can be produced from atmospheric carbon for zero net carbon impact

As I understand it, the production of complex hydrocarbons that can be used as a replacement for jet fuel is even more inefficient when compared to the production of hydrogen, as many of the known processes that start from CO2 actually use hydrogen in the process. So, although you could get energy-denser fuels, that would be at the cost of even more end-to-end efficiency.

Finally, hydrogen has a great energy / kg ratio, making it a very interesting fuel for aviation if the material science problems of hydrogen can be overcome, as it allows for longer distance flights / flights with more/heavier cargo.

That makes sense. I guess the questions would be whether the material science can really allow us to make use of that, and how the end-to-end efficiency weights against the infrastructure/engineering greenfield costs that synthetic fuels avoid. Being able to just keep using all existing engines/storage/distribution knowledge we already have is a major economic advantage. Electric will cover most land based transportation, and we need negative carbon technology anyway, so de novo hydrogen R&D has less to spread out across. If solar/nuclear becomes cheap enough, less efficient for more convenient might still be a worthwhile trade off. But I guess that's something for experts to crunch out. Thanks for the reply.

Don't forget that wind turbines still produce at night when there is less demand. So either you use that power to pump water upstream (if you have dams) or you can use it to produce Hydrogen.

There is also work underway on solar cells which directly produce hydrogen instead of electricity.

Or you can put it into a battery.

> As I understand it, the production of complex hydrocarbons that can be used as a replacement for jet fuel is even more inefficient when compared to the production of hydrogen, as many of the known processes that start from CO2 actually use hydrogen in the process.

Sure.

Here's the thing though: we can efficiently store those complex hydrocarbons. We can't efficiently store hydrogen. It destroys containers, and can escape any container we have devised(unless in liquid form?)

Many of the proponents of electrolysis claim that renewables will make energy a non-factor(or they'll propose nuclear power, etc). If that's the case, then add another step to capture CO2 as well.

> Finally, hydrogen has a great energy / kg ratio

Yes but energy per unit of volume is terrible.

The idea here is to do basically a cryoplane. Hydrogen can be generated on demand for the planes to refuel and can be stored highly compressed for that particular use case. We're not talking about using it for GA or cars.

As for volume - there's a reason why they call them cryoplanes ;) From Airbus - Liquid hydrogen is used as fuel for combustion with oxygen.

Liquified. So, low pressure. Aerogel insulation makes the tanks practical.

Volume on an aircraft is not expensive. Weight is expensive. Aerogel is very, very light. As is LH2.

Airports can make the LH2 directly on the spot from solar and wind power, so no need to transport or refine fuels.

Cryogenic fuels also give possible efficiency boost by precooling inlet air, but that's bigger gain for Sabre-style engines (HOTOL, Skylon, etc. SSTOs) or for efficient supersonic airliners (proposed derivation of Tu-144 used liquid H2 to gain low operating costs for it's typical Ma 2 cruise speed.)

> start from CO2

Why would that route even be taken though? Biofuel for jet engines is proven technology in limited use for many years now. Let plants deal with the CO2, since they're good at it, then process the resultant vegetable oils to produce a suitable diesel or jet fuel.

https://en.wikipedia.org/wiki/Aviation_biofuel

> Let plants deal with the CO2, since they're good at it

They aren't good with it at all. Convert CO2 slowly, take space to do that, take other resources like fertilizers, require some shielding from pests and herbs, can produce massive fires... They have advantages, but it's not a silver bullet, especially when you're in a hurry.

I believe popular proposals involve the use of algae. Regardless, farming in general is something modern civilization already has a strong competency in. Synthesizing jet fuel from scratch might make sense for the USN since they have nuclear reactors and all the water they want, but I think it's fair to say our civilization in the general case struggles with generating huge amounts of clean electricity. Plants do it with solar energy on a scale we should be jealous of. And so what of they take a few months to do it? They take that long to grow the food you eat too and it's not like you starve for months waiting for them.

Society uses absolutely mind-gobbling amounts of energy, mostly from fossil fuels. There's just not enough land available to produce even a significant fraction of that from biofuels. Never mind the ecological devastation that would be caused by converting the little wilderness left to intensive biofuel cultivation.

That is not to say biofuel don't have a role to play. Agriculture and forestry wastes, municipal biowaste, etc. can all be used for energy. And yeah, it makes sense to use the available biofuel for sectors that are otherwise very hard to decarbonise, such as aviation.

Algae, well, that has been researched for decades with very little to show for it. It has great potential, but it seems very difficult to make it actually work.

But this is my point; biofuel makes a ton of sense for aviation. Biofuels would be used in cases where there is no realistic alternative, particularly with airplanes. It's more practical and likely to succeed in this domain than hydrogen, and isn't a total pipe dream like battery powered planes. In other domains, like grid power or automobiles, solar/wind/etc are the way to go (I never suggested we run power plants off biofuel!), but there is still a fuckton of coal and natural gas plants that need to be replaced. Trying to also replace jet fuel with synth fuel at the same time will only set back efforts to close coal and natural gas plants. If we already had a surplus of clean electrical energy then things would be different, but we don't. Every solar farm dedicated to the production of synth fuel is a solar farm that isn't replacing a coal plant.

I just don't see any way that synthesis of jet fuel from atmospheric carbon makes any sense for anybody but a nuclear Navy.

The problem is twofold: The first is that biofuel doesn't scale. Maybe if we shifted all biofuel production to aviation we could do it. But the market is growing over the long term. Which means biofuels will eventually stop being a solution.

The other problem is that biofuel isn't actually all that green. It still needs lots of fertilizer, pesticides, farm equipment, etc., that ultimately gives it a relatively high GHG output.

At some point, we need to bite the bullet and go with hydrogen or one of its derivative fuels (e.g. ammonia, methanol, etc.). People are constantly trying to imagine a solution that avoids this outcome, but no one has come up with anything that looks like a viable alternative.

Maximum efficiency of 2%.

Even the worst electrolysis driven by cheap solar does better.

You've forgotten that we have a limited capacity for solar cell production and those solar cells are put to better use replacing coal plants than synthesizing something plants can synthesize.

Biofuel is just very, very expensive. The only reason it is used at all is heavy, heavy subsidies, often connected with corruption. (Cf. Archer-Daniels-Midland.)

And hydrogen or synth fuel aren't expensive? If these alternatives weren't more expensive then the problem would be moot because the industry would be voluntarily using them already.

They have always been more expensive than petroleum, and anyway would be made with power generated from petroleum or coal, or (in the case of hydrogen) from natural gas.

But that was before solar and wind became substantially cheaper than petroleum.

> Let plants deal with the CO2, since they're good at it,

This is never going to happen at the necessary scale. Where do you want to put all those plants? The planet's land area that is suitable for agriculture is already being used for food production. Well, apart from a few areas with natural reserves but you really don't want to touch those if you care about the environment. And if you don't then the whole clean energy topic is not for you anyway.

Sure some areas can be made a bit more efficient, but that's never going to be enough.

Biomass fuel production averages 1 watt per square meter compared to the 100 watts per square meter that solar power gives. Before fossil fuels civilization was in many ways energy constrained, forced to make harsh tradeoffs in how much land was devoted to forests for heating/iron production and how much to farmland for food.

Speaking generally about biofuel, it's a terrible idea - it's taking farming output and turning it into fuel, which is a bad idea because 1) it's unreliable and climate-dependent, and 2) we're basically just turning oil into plants and then turning the plants back into oil.

algae biofuels is what I'm hoping for.

This is a simplification, think about this:

You can manufacture hydrogen anywhere as long as you have water, an electrolyzer, and a power source. Oil can only be pumped out of the ground in certain places, water is technically _much_ more abundant everywhere.

Hydrogen storage tech has improved in the last twenty years. It's becoming affordable and "as safe" to store hydrogen at fuel densities similar to that of petroleum based fuels.

Obviously the devil's in the details, but shouldn't we attempt to engineer huge carbon emitters like container shipping or aircraft to run on hydrogen + solar?

>You can manufacture hydrogen anywhere as long as you have water, an electrolyzer, and a power source.

I don't understand what you're arguing with here? You can manufacture synthetic hydrocarbons anywhere so long as you have water, carbon dioxide, and a power source too. So I'm not sure why you see this as an advantage for hydrogen.

>but shouldn't we attempt to engineer huge carbon emitters like container shipping or aircraft to run on hydrogen + solar?

Again, why hydrogen from solar vs hydrocarbons from solar? From a global warming perspective what matters is net GHG.

> You can manufacture synthetic hydrocarbons anywhere so long as you have water, carbon dioxide

You need coal or some biomass to manufacture synthetic hydrocarbons. And tail emission is still a concern.

You can extract carbon from atmosphere.

Organic reactions to produce complex molecules have an horrible efficiency. (For example, photosynthesis in plants has a 1% or 2% of efficiency.)

Perhaps the efficiency to produce methane is not so bad, but a chain with 6 or 7 carbons like gasoline is very complicated. It is not equivalent if they are aligned or they form a ramified structure like a tree, and it is very difficult to choose which one they will form unless you have a very complex process. There are some graphics in https://en.wikipedia.org/wiki/Octane_rating#Isooctane_as_a_r...

> Perhaps the efficiency to produce methane is not so bad, but a chain with 6 or 7 carbons like gasoline is very complicated.

The Fischer-Tropsch process is almost a hundred years old, and has been used on an industrial scale. E.g. Nazi Germany used it to produce a non-trivial fraction of their liquid fuels during WWII. Similarly South Africa used it during Apartheid. Now FT itself is better suited for producing mostly straight hydrocarbon chains like diesel or jet fuel rather than the branched ones seen in high octane gasoline.

The efficiency of traditional FT is around 40%, but by adding extra hydrogen to the process it can be increased by around a factor of 3 (the efficiency meaning the energy content of the carbon carrying feedstock vs the output, so over 100% is possible as hydrogen provides some of the energy in the final product).

The problem is to get the CO needed for FT in a carbon-neutral manner. Biomass is limited, capturing and reducing CO2 is expensive and requires a lot of energy.

"The Fischer-Tropsch process is almost a hundred years old, and has been used on an industrial scale. E.g. Nazi Germany used it to produce a non-trivial fraction of their liquid fuels during WWII."

I know it's not an argument against, but the Nazis lost WW2 because they ran out of oil. They invaded Russia at the last moment they could before they wouldn't have the oil to invade, and when they invaded they made a beeline for Russia's oil fields.

They were desperate for fuel from anywhere, because their tanks are useless and they either have to fight to the death wherever their tanks run out of fuel, or abandon their tanks.

...also, the main non-fossil fuel the nazis used was probably hay. For their horses. It's just a super weird example.

> I know it's not an argument against, but the Nazis lost WW2 because they ran out of oil.

Well, they would most likely have lost anyway, but failing to capture the Caucasian oil fields as well as the Allies attacking their fuel infrastructure certainly hastened their demise.

One can make a good argument that the strategic bombing on the western front in WWII was largely ineffective, but there were parts that were wildly successful, like attacking the fuel infrastructure.

As for why they didn't just use synthetic fuel all the way instead of just a fraction of the total use, I don't know. I guess they just couldn't scale it up fast enough. And certainly getting oil from the ground is much cheaper than creating synthetic fuel from coal in huge industrial parks.

> ...also, the main non-fossil fuel the nazis used was probably hay. For their horses. It's just a super weird example.

Sure. Blitzkrieg captured the public imagination, but in reality only a small fraction of the German army were these mechanized shock troops. Most were infantry that moved by train or marched with horses carrying equipment.

> It's becoming affordable and "as safe" to store hydrogen at fuel densities similar to that of petroleum based fuels.

Have we solved hydrogen embrittlement issues? What about hydrogen leaking out of containers?

Hydrogen leaking is an issue for longer term storage, not particularly an issue for liquid hydrogen that is going to run an airplane for a very specific number of hours.

> for a very specific number of hours

Unless you plan on throwing away the planes after one or two flights, those tanks will be filled with hydrogen for tens of thousands of hours over their lifespan. Perhaps they have light weight alloys that can cope with this though, I'm no metallurgist..

I never addressed the lifespan of containers, that's not same as hydrogen leakage.

Still embrittlement is an issue. How long until the hydrogen tank needs replacement?

Liquid per liquid, why not liquid methane that's far easier to work with and can be manufactured at zero carbon emissions.

Hydrogen embrittlement leads to hydrogen leaks. Unless there are advanced alloys that resist it, any aircraft part that comes into contact with the hydrogen will be subject to accelerated 'wear'.

Frankly hydrogen planes are nonsense. Even if it works on paper, it's not realistic to expect all the extant planes to be rapidly phased out with brand new planes with exotic designs that aircraft manufacturers, pilots, and airlines don't have any experience with. On the other hand, old planes can be retrofitted to burn biofuel and new production planes of old proven designs can be made to burn biofuel from the get-go.

'Concept planes' like the ZEROe are basically just science fiction meant to wow shareholders and give the public the impression that legislative reform isn't necessary because corporations are already taking the problem seriously. I don't think these are serious proposals, it's just corporate propaganda. Car companies pushed hydrogen cars for years for the same reason. Hydrogen fuel cell cars were never going to happen for a myriad of practical reasons and car manufacturers knew it but promoted the concept anyway for PR reasons. ZEROe is Airbus doing the same.

Not to mention hydrogen fires being invisible to unaided eye in daylight.

Hydrogen at atmospheric pressure is hard to ignite in uncontrolled circumstances. It wants a specific gas/air ratio.

Thought experiment: how close would the Hindenburg have been to missing that specific ratio?

The thing is, no matter how far you are overshooting or undershooting that ratio, as soon as you are doing both there'll be a boundary where it's matching.

The Hindenburg was blown up with a bomb. The flames in the film are kerosine. Trotting out the Hindenburg weakens your case, because no H2 lofted dirigible suffered an unassisted ignition event.

The boundary, if there is any, also has to coincide with an ignition source. Furthermore, the ignited bit needs enough fuel around it to build a chain reaction.

Systems will be designed to maintain positive airflow to keep concentration, in the event of a leak, below the threshold. That is engineering.

Might be space flight specific (ideal ratios/big flows of hydrogen maybe ?) but NASA seems to take hydfrogen fires very seriously: https://spinoff.nasa.gov/spinoff1997/ps1.html

> engineer huge carbon emitters like container shipping or aircraft to run on hydrogen + solar?

Why hydrogen at all then? The source is solar so let's keep it at that, why the middle state?

Really? You think solar panels on the wings would provide enough power to fly?

Aircraft need concentrated energy. Batteries are nowhere near enough for anything big. Liquid H2 is the best possible fuel for aircraft, once certain practicalities are worked out, and can be produced from solar and wind.

OK So this is just about conversion.

H2 doesn't need solar, it needs power, it doesn't care where that power came from. It's essentially a disconnected system, to sell it as if H2 somehow means solar (which I get from earlier threads) is a little bit disingenuous.

Not even a little bit. Wind and solar are the cheapest source of power, with their usability reduced only by surge demands. Liquifying H2 would absorb peak production and be unaffected by temporary lulls.

Unless you think it would be natural to use the more expensive power instead?

That makes much more sense, was not thinking in terms of surges.

> The infrastructure has no multiple use, unlike improvements to the grid.

You need hydrogen for steel production, ammonia, and synthetic hydrocarbon production. Likely cement, glass, and plastic production too.

> So why not just do full manufacture of regular fuel? I've never understood where hydrogen fits between improving batteries/capacitors and synthetic hydrocarbons.

Because fuel cells are batteries themselves, and synthetic hydrocarbon needs hydrogen.

> So why not just do full manufacture of regular fuel?

Always fascinating when people use "just" in sentences where they propose something that hasn't been done before (at least not at any scale that would matter) and that is a huge challenge.

As for the reason, my understading is that producing fuels from electricity is just very inefficient. You still use hydrogen as the first step and then add other steps to produce hydrocarbons. Using hydrogen spares a few conversion steps and you'll need less electricity (but likely still a lot).

"Just" is a great example of a word carrying more subtext than text. "It's just a stabbing."

An advantage of hydrogen is that it can be used in fuel cells, producing only water as reaction product. Using hydrocarbons of any sort in existing engines leads to the production of nitrogen oxides due to the presence of nitrogen in the air.

Thanks for this reply as well. But batteries will cover most land use, and going by sources like the EIA [1] the primary source of anthropogenic N2O emissions is overwhelmingly agriculture, at 70+%. Other sources indicate that may have fallen somewhat since then, but even so everything I can find indicates fossil fuel combustion is still only around 10%. And of that air travel will be single digits. Not nothing, but I'm not sure it'd justify massive effort alone.

I don't think hydrogen is a good energy storage solution. I do see its advantage over hydrocarbons for certain applications but the difficulties in storing hydrogen - a small molecule which is sort of an escape artist - make it less than ideal for many purposes. Fuel cells can get contaminated unless they're run with pure hydrogen and oxygen. In applications where hydrogen is already available - e.g. H₂-LOX fuelled rockets - it makes sense to use fuel cells. It might be possible to run aircraft using H₂-fed fuel cells if these can be made powerful and light enough to fit the application, the advantage over battery-electric here is that the energy density of liquid hydrogen is much higher than that of current battery technology. For the rest - cars being a good example - I think batteries are currently a better solution.

Large scale energy storage is probably handled better using either redox flow batteries [1], hydro storage systems or energy storage vaults using concrete blocks which are lifted when there is too much energy available, lowered when energy is needed.

[1] https://en.wikipedia.org/wiki/Flow_battery

Manufactured fuels have two main economies: one is energy input per energy output (usually bigger than one and the goal is to get that number as close to one as possible) and the other is transformer system cost, particularly cost that even applies when idle. If this number is very high fuel manufacturing won't be opportunistic at all and will keep the conversion systems at high utilization, offering input energy prices as needed and selling at prices depending on demand. A very low investment/idleness cost on the other hand would lead to wildly variable utilization depending on output demand and input availability.

Hydrogen is usually considered a very simple conversation (e.g. compared to synthetic hydrocarbons) that would be very open to opportunistic transformation utilization.

"Only fire up the converters when there's an open order for synfuel and the grid has excess power", that's the dream. I'm not convinced that this dream is significantly more viable for simple hydrogen than for more complex synfuels (particularly factoring in the many ways hydrogen isn't easy to store), but I can easily see how some people might want to believe that.

With the advent of cheap aerogels, LH2 is much easier to store than it once was.

It's difficult to tell how they will deal with hydrogen as they don't even try to address all the points you have made. The articles available just say "hydrogen stored in tank under the wing". Seems investors in green tech. will be betting on both technologies aka having facilities that can both produce hydrogen and synt. fuels.

https://www.energyglobal.com/special-reports/28082020/haldor...

https://www.engineering.com/ElectronicsDesign/ElectronicsDes...

Well the answer to that is that it is not that simple. We know how synthesize hydrogen and scale it and what it will cost because companies are already doing that today. Synthesizing other fuels may be long term feasible but are simply not commonly done right now. More importantly, there don't seem to be many companies planning to do that right now. Airbus is not in the business of producing fuel so it would need such a company to step up and do it before they can start designing products for it.

E.g. liquid methane would be a good alternative for hydrogen. The appeal of both hydrogen and methane would be that converting engines to burn it seems doable and given that rockets can use either as propellant there is no question about their suitability as a fuel from an efficiency point of view. The difference of course is that with methane it would end up coming from fossil reserves which from the point of view going carbon neutral kind of defeats the purpose of making these planes carbon neutral. Swapping out kerosene for methane kind of just moves that problem. IMHO it's still worth exploring (and some companies have actually done that). Also, I imagine synthetic methane is eventually going to happen and kill the market for natural gas. Just not in the next decades.

The key here is a combination of becoming carbon neutral and doing it in a cost effective and timely way. Synthetic methane may or may not actually become a thing but the simple reality is that it's an industry that simply does not currently exist. Clean hydrogen production on the other hand exists today (at small scale) and seems to be on a feasible path to scaling in the next 15 years or so. Additionally hydrogen infrastructure is something that is attracting lots of investment currently. Airbus is targeting a launch in the mid to late 2030's. So they need to be targeting a fuel source that they can reasonably expect to have some supply for by then so they can actually fuel the planes where-ever they need to fly. So the supply and infrastructure for hydrogen are not a certainty but at least on a path where it could be there in 15 years. The rest is just solving engineering problems. They seem to believe they can do it.

Getting the carbon is a huge problem. Atmospheric CO2 is at something like 400ppm - that's 0.04% of air. Given that most hydrocarbons contain a fairly large amount of carbon, that means you have to process a huge amount of air to get the CO2 needed for your fuel. That's before all the other process inefficiencies.

I agree hydrogen isn't the best fuel, but the round trip efficiency of any other synthesised hydrocarbon must be an order of magnitude lower - and even if solar energy is abundant, it's not so abundant we can waste 90%+ of it on process inefficiencies.

> The infrastructure has no multiple use

Hydrogen solves long term storage for electricity generation and heating (winter time)

https://model.energy/

It sounds futuristic though and the exhaust is water!

Reply on news.ycombinator.com