Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Ask HN: Is hydrogen likely to be a major source of power in the next 10 years?
138 points by eigenvalue on Sept 23, 2020 | hide | past | favorite | 211 comments
There has been hype around hydrogen fuel cells being a source of green power for decades, but the hype cycle is currently ramping up significantly. Hydrogen fuel cell stocks such as PLUG and FCEL have gone up 10x despite the lack of any significant new breakthrough technology and a history of loss making and value destruction.

Just today, Bank of America research released a 103 page "Hydrogen Primer" in which they predict that the hydrogen space will generate $2.5 trillion in direct revenues and $11 trillion of indirect infrastructure potential. They believe that a tipping point is coming soon because of the falling cost of renewable energy and electrolysers used to produce "green" hydrogen, as well as better efficiencies in fuel cells.

My question is for the experts in the crowd here, either based on engineering experience or on first principles and physics, does this seem likely to you? Why do we really need hydrogen? It seems we are getting to the point where wind power and particularly solar power are now cost effective. Once we have better batteries for storage, what problem do we have left that requires hydrogen to solve? Is this just a giant promotional bubble being pushed by Wall Street and unscrupulous companies trying to sell a dream? I may sound skeptical but I genuinely don't know and would like to hear from people with more expertise.



Hydrogen fuel cells are not a source of energy, since hydrogen (H2) is not naturally found on Earth. All H2 must be generated in a process which consumes either natural gas or eletricity. Therefore H2 fuel cells are more aptly described as an energy storage system than as an energy source.

Therefore the relevant comparison is between H2 fuel cells and Li ion batteries. Batteries have much higher full cycle efficiency (energy input -> storage -> energy output), but they are large and heavy. H2 can be stored in less volume and less weight, but it is less efficient. In my opinion, batteries will be a technically and economicly superior solution for all uses other than where weight is extremely critical such as in aircraft.


> In my opinion, batteries will be a technically and economicly superior solution for all uses other than where weight is extremely critical such as in aircraft.

I was excited about the use of hydrogen in aircraft for a while but the more I looked into it the less likely it seems. I think the advantages of liquid fuels are too great and that even if there were no oil production it would still make more sense to synthesize jet fuel or some other liquid fuel for use in aviation.

This has nothing to do with feasibility (the Soviet Union had an airliner that ran on hydrogen, so hydrogen aircraft are absolutely feasible), just with lock-in effects and ease of use.


It's not just liquid vs gas.

Hydrogen has a couple of "unique" properties that make it especially disadvantageous.

Namely: It'll diffuse through just about anything on a fairly short time scale if it isn't kept as a liquid at cryogenic temperatures, and it causes almost all metals to become brittle.

I think hydrogen is basically a dead-end as energy storage, except in a few niche applications where it's produced as a byproduct and can be consumed immediately.


Steel production, where the hydrogen is a chemical reagent, might be one of the few exceptions where hydrogen is really the best choice.


Not to mention liquid H2 has quite a bit lower energy density than liquid fossil fuels. Many rockets have used kerosene in their first stages for this reason. Not the end of the world for aircraft, but certainly another major disadvantage.


But the energy density per mass is still better than Li-ion batteries.


Li-ion batteries are improving at a faster rate, and you can use your gravitational energy to recharge the battery.


How does that work? You run your propellor in reverse like a wind turbine?

This doesn't sound completely ridiculous but is there even any proof of concept for that? The gravitational potential energy doesn't seem all that high - most airplane fuel is burnt maintaining height, not gaining it - to justify the complexity of recovering some small fraction of it. Still, I once thought the same about regenerative braking in electric cars.


Hydrogen is a storage mechanism; there'd be a way to reclaim your energy either way


It's also prone to explode, and is a potent (indirect) greenhouse gas if it leaks into the atmosphere. Agreed that it will have some niche applications though. As well as the byproduct situation, there are likely situations where batteries wouldn't be feasible due to insufficient capacity or too great of weight. Aerospace maybe?


Hydrogen (and helium) are too light to stay in the atmosphere and are lost once released.


How is hydrogen a greenhouse gas (and what do you mean by "indirect") ?


I think it'd be an "indirect" GG because it's electricity intensive to obtain it, and that likely comes from a carbon emitting source. The same argument is often made about electric vehicles.


In this case it's about how it interacts with OH radicals in the atmosphere, meaning other GHG stay in the atmosphere longer. So it's not a greenhouse gas, but it makes others hang around longer.

> The reaction of hydrogen with OH radicals has a further side-effect of reducing the availability of OH radicals with potential impacts on the build-up of greenhouse gases.

https://assets.publishing.service.gov.uk/government/uploads/...


Well you don't typically get hydrogen from H2O, you get it from CH4. So your byproduct isn't oxygen, it is carbon. Unlike electric vehicles, it doesn't get greener as the grid changes to wind/solar/nuclear/hydro. There's just more hydrogens per hydrocarbon than water. Not to mention dealing with oxidation.


> Namely: It'll diffuse through just about anything on a fairly short time scale if it isn't kept as a liquid at cryogenic temperatures, and it causes almost all metals to become brittle.

Both claims are wildly wrong. Neither actually happens except at high temperatures and only with certain metals.


Could you elaborate about “wildly wrong”?

I am not an expert on this, but random internet searches such as [0] indicate that common materials, like high strength steel, are susceptible to both diffusion and embrittlement at room temperature

[0] https://www.energy.gov/sites/prod/files/2014/03/f10/pipeline...


Those tests seem to be at >200˚C if I’m not mistaken. Hardly room temperature. Regardless, the main trick is to avoid using certain types of high-strength steel, line exposed surfaces with polymers, or using composites where ever possible. The problem is for the most part solved.


So you're admitting the problem exists, which is my entire point.

It's hardly a theoretical-only problem, either:

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


In the same sense all chemicals have handling issues. In practice, it is basically solved.

That's probably due to them using a vulnerable alloy, and was likely exposed to hydrogen while at a high temperatures. This is nothing like a gas tank filled will hydrogen.


What's a high temperature though? -50C?


Well into the hundreds of degrees C.


That's why some bet on ammoniac.


It astonishes me how far behind aviation is behind surface transport when it comes to sustainable fuels. They still use leaded gas!

It seems you would try methane before you try hydrogen:

https://energynews.us/2013/08/26/midwest/could-natural-gas-f...

Petroleum-based liquid fuels contain mixed entity hydrocarbons that match a specification; if you have the perfect feedstock and markets for the fractions you don't use this is cheap. Worst-case you have to synthesize them with Fischer-Tropsch chemistry

https://en.wikipedia.org/wiki/Fischer%E2%80%93Tropsch_proces...

which is such a PITA (normally a chem E would be delighted that you can use iron as a catalyst, but it works so poorly that a desperate search of the other 91 elements found just 2 that sorta-kinda work)

The issue there is you are sticking C's onto the end of a chain and you have to deal with a network of reactions that produce products you couldn't care less for, such as petroleum jelly that gums up your catalyst, sticks up your downrisers, etc.

Sustainable motor fuels tend to be single entities such as 1-butanol, dimethyl ether, etc. You might need to blend something in for low temperature starting, but it's possible for a single entity fuel to be synthesized with decent yield.

General av is seen as a backwater that is barely hanging on and couldn't possibly get the lead out. The USAF has done biofuels trials, but commercial av is spooked to try anything that could leave passengers up in the air.


> It astonishes me how far behind aviation is behind surface transport when it comes to sustainable fuels. They still use leaded gas!

It's important to qualify this: jet fuel contains no lead, and constitutes the vast majority of commercial aviation, and hence of emissions.

Jet fuel itself is kerosene, with some additives that aren't superb in raw form, but nothing so durable in the environment as lead.

A sibling comment does a fine job of explaining why avgas still contains lead, and to be sure, a plan to phase that out except where absolutely necessary would be welcome.


> General av is seen as a backwater that is barely hanging on and couldn't possibly get the lead out.

The issue is infrastructure. Modern engines run fine on unleaded fuel, but the little airstrips in the middle of nowhere all have 100LL and/or jet fuel, so that's what pilots use.

I know a few pilots who are prefer to use unleaded in their Rotax engines to prevent spark plug fouling, and they have to jump through some pretty ridiculous hoops to get gas, while their buddies just fill up with whatever is on the field.


General aviation fuel is leaded but jet fuel does not contain lead.


The issue is that while cars only have ~15 year lifetime, small general aviation aircraft have ~50 year lifetimes.


Not particularly relevant to the lead debate though as the engines are replaced after 1,500 - 2,000 hours and retrofit Diesel engines running Jet A are available on lots of aircraft (but people don’t buy them as they’re expensive and heavy)


The problem with LNG is that it's still a fossil fuel, still cryogenic, and that methane itself is a potent GHG. The first problem you could solve using synthetic methane, but the latter two still drives you towards LH2 as being the better option.


Why not both? ;)

Airbus is experimenting with liquid hydrogen: https://www.airbus.com/innovation/zero-emission/hydrogen/zer...


A lot of the concepts for hydrogen aircraft use liquid hydrogen though.


It's hard to store, transport, and handle though, so I don't think it has much of an advantage.


Synthetic jet fuel will need hydrogen as an input chemical. This is not an "either-or" question.


I didn't say we don't need hydrogen though. I just said it makes more sense to use liquid fuels rather than hydrogen in aircraft.


That falls under the “maybe” category. Synfuel are not easy to make, and will add a substantial layer of extra complexity to an already complex process. If it can’t be done cheaply then it might not happen or make sense. Hydrogen is not an alien substance BTW. It’s possible that we figure out hydrogen before synfuels get cheap.


Even in rockets, where weight is extremely critical, SpaceX chose to use kerosene/methane due to the problems of working with LH2 (very low density, very cold temperatures, hydrogen embrittlement).

Depending on energy density and power density requirements, I think a combination of batteries and traditional fossil fuels will win in the short term. In the long term, when we have stopped burning fossil fuels for electricity, then we can use electricity to manufacture them. Under this path, hydrogen will have limited importance.


Preventing explosions in the event of H2 leaks is difficult too. H2 has one of the lowest ignition energy curves of any gas. Electronics used in areas with H2 must follow very strict regulations (worse than medical requirements in many ways).


The explosion happens when you cross the ignition energy threshold. So having a low ignition energy threshold can actually be an advantage as then much less fuel is available to drive the explosion.

Agreed that doing intrinsic safety for H2 is hell. I did once design a LED sidelight for the relevant class with significant current (redundant electronic energy regulation). So it is possible to do stuff, it just takes more work. The tendency for modern electronics to be very low voltage helps a lot.


More likely: A small amount of the leaked hydrogen encounters an ignition point, igniting the primary mass.

The ignition zone is larger, risk is higher.


Pure hydrogen can’t burn without oxygen, so you generally get a tiny flare and then a flame based the size of the crack. It’s really about the same risks as using gasoline which almost never causes a significant detonation.


With vehicle, heating, rocket, or aviation applications, quantities involved are nontrivial. Hydrogen's flammable and explosive ratios with oxygen are large, and contrast starkly with other fuels; kerosene (jet/rocket fuel), bunker oil, and deisel, which ignite with difficulty, petrol, which ignites readily but not explosively, and even natural gas which deflagrates rather than explodes under most circumstances.

Add in hydrogen's extreme tendency to leak, metal embrittlement, and high pressures and/or low temperatures, and the risks are immense. Particularly at scale, in widespread use, with poor maintenance and inspections.


It’s technically explosive in many situations where the location or net energy released makes that irrelevant. Being significantly lighter than air you might hear a loud bang above the vehicle where gasoline’s foof ends up being significantly more deadly.

It’s not strictly better or worse, just different. Diesel for example can be very dangerous in a large open topped container, hydrogen just doesn’t stick around in that environment.

PS: Consider what it would take to make a large hydrogen fuel air bomb that’s as effective as it’s hydrocarbon equivalent.


> Even in rockets, where weight is extremely critical, SpaceX chose to use kerosene/methane due to the problems of working with LH2 (very low density, very cold temperatures, hydrogen embrittlement).

That and they have very few missions where the increased ISP from LH2 would actually benefit them. The other upside of LH2 and LOX is that you can run it in engines, but you can also run it in fuel cells directly and generate a surprising amount of electricity from a very small package -- the other benefit there is that the Fuel Cells produce pure water as a byproduct.

The Space Shuttle had three fuel cells which provided more than enough power for a full mission. Most missions could be run on two without any compromises. Also, when it was visiting the ISS, it was a convenient means of supplying water to the station.

On non ISS missions, so much water is generated that they have to dump it overboard.


> Hydrogen fuel cells are not a source of energy, since hydrogen (H2) is not naturally found on Earth. All H2 must be generated in a process which consumes either natural gas or eletricity. Therefore H2 fuel cells are more aptly described as an energy storage system than as an energy source.

100% this.

IMHO, this is by far the biggest misconception about hydrogen among people.


Not actually true. H2 is found naturally, although we don't have an good assessment how much exists or whether there are significant deposits.


To my understanding, natural discoverability of H2 gas on earth surface is very low due to the gas being very light and buoy itself up the atmosphere, possibly escape the planet entirely.


Hydrogen is being created and consumed on a vast scale on Earth. The possibility that there are huge pockets of hydrogen gas trapped under impermeable rock layers cannot be ruled out. For one thing, natural gas pretty often contains hydrogen as a constituent.


But current battery technologies need elements that require mining and other environmentally taxing processes. Recycling batteries can be energy consuming too. With hydrogen the cost structure of the whole fuel lifecycle is much clearer. Hydrogen also depletes along the trip unlike battery which will be massive dead weight for planes & ships.


FWIW I upvoted you just because the question is reasonable, but I think you're wrong in balance. Last I looked all the current processes for making hydrogen are in fact more environmentally taxing (and worse, CO2 emitting) than the one-time cost of mining lithium.

A LiOn battery is good for many many cycles, and can then be recycled. Hydrogen must be produced and then consumed, and is usually produced from natural gas, itself a greenhouse gas that is on the main a byproduct of the oil industry.


Green hydrogen is the long-term option, which makes hydrogen the much less resource demanding option. People who keep on repeating the 'hydrogen = fossil fuel" argument is just the same kind of argument leveled against battery powered cars such as "coal powered cars."


CO2 emitting but it's point source so straightforward capture and sequestration.


Yes, the cost structure is very visibly more expensive. The problem is the huge amount of additional energy needed to generate the hydrogen. Hydrogen storage is also very expensive. The weight reduction is pretty neglectible, as hydrogen is the lightest element in the universe.


For anyone wondering what kind of numbers we're talking about in practice -- the Toyota Mirai holds about 5kg of Hydrogen.


In a pair of tanks that weigh 87.5 kg.


That's not that bad right? 15 gallons of gas could be 43 kg, and the gasoline fuel tank has some weight too.

Although, after google search, I see that the Mirai weight is 4075 lb, and the Prius weighs around 3100 lb.


That ratio looks depressing, but what would equivalent batteries weigh? 500 kg? 1,000 kg?


A Tesla battery pack of comparable range is about 500kg.


Note that the mirai also needs the fuel cells, which adds another 70 kg, so the difference isn't as stark


Once we start that game, we can also add the weight of the Mirai's 1.6kWh NiMH battery and other esoterica.

The proof is in the pudding: the curb weight of a Mirai is more than that of the larger and more capable Model 3.


How important is weight in their designs though? They're cars, so they don't have to hold themselves up, and they probably both have regenerative braking


You are right, weight isn't extremely important, especially with regenerative braking. It is only interesting in the comparison with battery electric cars - if the hydrogen car is even heavier, then there is basically no advantage left for the hydrogen car.


What is the energy came free, from renewables? What would the cost structure for hydrogen be in that case?


You'd still have to compare it to the equivalent usage of the same energy into a LiOn battery, and in many cases it wouldn't fare well.

Maybe there's a future world in which we can do it all in orbit -- harvest comets, produce hydrogen in space using solar power or nuclear, and efficiently drop it down to earth in some safe way. That would definitely alter the equation. But by the time we figure that out, it'll probably be too late.


If we're talking about space mining is Lithium abundant too? It's number 3 so should be everywhere in the universe?


Cost structure: ~$5/kg now; forecasted to drop to $2/kg by 2030 and $1/kg by 2050.

https://about.bnef.com/blog/hydrogen-economy-offers-promisin...


There is no free energy from renewables. Renewables are cheap, but not free. And we are very far from even having a 100% renewable grid.


When the spot price for energy can go negative due to renewable over production it's fair to say that energy can be free or even better.


Yes this is because there isn't enough grid storage yet. Massive amounts of stationary storage need to be produced whether that be pumped hydro or lithium ion battery or some other means of energy storage.


I recently worked out that the UK's installed capacity of pumped hydro is around 28GWh. This is around the worldwide total production of lithium batteries per year circa 2016 (realistically it may have doubled since then).

Personally I don't going into competition with laptops, mobile phones and car battery manufacturers as a viable long term strategy for grid scale storage.


I think your numbers are outdated, that's about the production numbers of Tesla alone, not counting the other manufacturers. And Tesla just announced their plans to move towards 3 TWh/year production rate towards 2030.

Any electric car plugged in is a great buffer for surplus electricity production.


In theory yes, in practice, which electric car charger has a built in 3kw inverter for backfeeding? Have you seen how large a 3kw inverter is? They are also not cheap. Secondly, most homes aren't metered in both directions. What is the ROI of one car plugged into the grid in this system? How much does it depend on how long the vehicle is actually plugged in and whether it is plugged in at useful times.

Taking the number above for installed capacity of pumped hydro. You are looking at nearly half a million cars and homes to match the capacity (assuming 50kwh per vehicle), before considering the plugged in factor.

At say a nice round £1000 to upgrade each home you are already looking at a cool half a billion. Already around the cost of the Dinorwig pumped hydro station before you figure out how to compensate electric car owners for the extra wear on their batteries or the fact that when they leave for work their car is flat because it has been powering your neighbors electric showers.


Yes, in those rather short time periods, energy prices are free. But for those periods, batteries are the better storage solution, as the total amount of "free" energy is rather small. Electrolysis plants would have to run basically 24/7 to be viable. More battery electric cars on the roads also mean a lot of storage for excess electricity. Just use the top 10 to 20 percent of the capacity to absorb production peaks.


I think its not as bad as you make it out to be. Any battery that will be used in flight might have nickel in it but that's about the only thing that's problematic. The anode will likely be silicon.

Pretty much all battery research is targeting materials that are common. The DoD made a large bet on Lithium-Sulfer batteries.

Battery recycling can be energy intensive if you go for the full metallurgical extraction. But pretty much all car manufacturing plants want to just 'disassemble' the battery and use other far less energy intensive processes to get battery grade materials out. Also, at least in the west, these factories are generally located in location where there is cheap green energy or at least there is the potential. That is both Tesla/Panasonic and Northvolt (Europe) plan.

With batteries you can reload the battery from your gravitational energy. That is actually quite a significant range extender but it might be a regulatory issue.


> But current battery technologies need elements that require mining and other environmentally taxing processes

But hydrogen requires production, either from fossil fuels, or electrolysis, both can be environmentally taxing. It requires transport. It requires either liquefaction or compression. It require mining to manufacture suitable storage tanks - which will become brittle and have to be replaced pretty often.

> Hydrogen also depletes along the trip unlike battery which will be massive dead weight for planes & ships.

This is indeed a major difference.

Frankly, if you want to burn hydrogen, you're better off synthesizing something like methane. It's a far more effective storage mechanism. After all, if you are willing to spend the massive amount of resources required to do large-scale electrolysis, you might as well go the extra step and setup some sabatier reactors. You'll recapture the CO2 too.


Does "less efficient" mean less energy dense, or our current methods for converting its energy to electricity aren't good?

If the former, then what does "H2 can be stored in less volume and less weight" mean? I would assume you would want to compare the mass required to store the same amount of energy, in which case it sounds like H2 takes up more space and weight than Li-ion?

If the latter, then isn't it just a matter of improving our technologies? Or is there some law of nature that makes efficient conversion to electricity impractical?


It’s admittedly been a while since I’ve looked at it, but I recall that the creation of H2 using existing processes takes quite a bit more energy than you ultimately extract back from it. For example, electrolysis of H2O to H2 is a pretty lossy process (I think around 60-70% of the input power is used to break the H2O bond and the other 30-40% is released as heat).


I'm a total noob in that area, but I always thought that hydrogen was super efficient when converting it from water + oxygen and then back. Why battery is better?


Roundtrip efficiency for batteries are 70-95% vs 47% for H2.

https://www.sciencedirect.com/topics/engineering/round-trip-...

Roundtrip efficiency is not the reason to use hydrogen for anything.


Do you know what’s the (current practical and maximum theoretical) roundtrip efficiency of hydrocarbons (gasoline, alcohol, methane, ...)? Could be a better alternative than H2 (and longer lasting than batteries, even if lower efficiency).


Elektrolysis isn't very efficient, neither are fuel cells, when you compare it to batteries. On top of that you have the energy usage for transporting and most of all compressing the hydrogen for storage.


Bear in mind -- a fueling infrastructure includes a lot of potential points of inefficiency outside of the chemical conversions that are taking place during generation and consumption.

Your fuel of choice also must be transmitted from the point of generation to the point of consumption. For electricity, this is easy, you can run it down a wire. For gasoline, you need a tank that can hold liquid. For hydrogen, you need to cool it to −253°C and compress it in a cryogenic tank.


And water scarcity is expected to be a growing issue for many parts of the world. I guess we could use sea water, but then the shipping you mention comes into play.

Just a side note, you could ship it in hydride tanks.


The weight factor also makes it easier to transport. In off-grid use-cases, it could make a lot of sense to get hydrogen delivered instead of setting up solar/wind + battery.


Not really.

Again, it's kind of stuck in the middle: energy density by volume sucks compared to hydrocarbons, and if it isn't cryogenic, it really sucks, and cryo is expensive. Transporting fuel by truck is volume limited, not weight limited.

If we're comparing to batteries, then... ok, but you kinda only have to move the batteries once, and then you can either recharge on site with solar or connect to grid. With hydrogen, you have to keep bringing it, and you may as well have hydrocarbons at that point.

The lack of carbon emissions on-site is what makes it tempting, but it's just easier to work around this problem in various ways than it is to pay the hydrogen penalty on an ongoing basis. To a first approximation, all commercial hydrogen is cracked off of methane anyway, and electrolysis is a pretty inefficient way to use renewable electricity, so even that just displaces emissions to fossil fuel plants unless none of those exist anymore.


Not quite.

Usage wise: Batteries have lower gravimetric density than h2 tanks. Batteries have about the same gravimetric density than h2.

Economic wise: Batteries have cobalt which is more scarce than oil/nat. gas from which h2 is made.


In theory you can also transport hydrogen through the pipes used for gas. I personally think you’ll run into problems with leaks really quickly but who knows.


Hydrogen gas has the smallest size of a molecule which is why only pyromaniacs might try that.


Right. Hydrogen can be used to store energy.

Unlike flowing water which may turn a generator, hydrogen isn't a continuing source of generating power.


do you have a book recommendation for these kinds of things maybe?


That's why Astroid Mining and inter-steller mining technologies can't come soon enough.

I can't believe that we went to the moon in the 1960's and now we cannot get support 60 years later for trying to economically exploit space.


What can we mine in space that we can't mine more cheaply on Earth?


Unobtanium


It’s environmentally friendly to mine asteroids, so in a way, it will always be “cheaper” to pollute an unlivable rock rather than our only sustainable habitat.


It'll take like 100 year to get to that point. That's 100 yeard of space engineering simply polluting Earth and then maybe another 100 years to break even?


Yes, when the space infrastructure is there.

Once in orbit, it takes very little energy to go to an asteroid and to launch that asteroid to a parking orbit compared to dredging mountains of soil and rock on Earth. The yields would be extremely high; there's more gold and platinum in some Inner Belt asteroids than we've ever mined on Earth so far, and it's just floating there for you to grab and bite chunks out of.


> Once in orbit, it takes very little energy to go to an asteroid and to launch that asteroid to a parking orbit compared to dredging mountains of soil and rock on Earth. The yields would be extremely high; there's more gold and platinum in some Inner Belt asteroids than we've ever mined on Earth so far, and it's just floating there for you to grab and bite chunks out of.

This is not true at all. This is laughably absurd. Getting from Earth's surface to LEO takes about 9.8km/s of delta-v but then getting from LEO to another destination requires more delta-v. To get to the Moon requires another 6km/s of delta-v on top of the 9.8km/s you needed to get from the ground into orbit. There's a reason the Saturn V was a giant rocket with a teeny tiny capsule mounted on top. Getting places in space requires a lot of propellant.

Here's a nice list of NEAs and the delta-v required to reach them (one way) from LEO[0]. The list is very helpfully sorted by delta-v requirements. You'll notice that the easiest NEA (2018 AV2) requires an additional 3.758km/s of delta-v. The easiest to reach NEA on the least requires more than half the delta-v that it would take to land on the Moon. Once you rendezvous with an asteroid you need to come home which requires yet more delta-v (thus more propellant). Moving an asteroid to orbit Earth is not in any way as simple as you might think from reading science fiction.

So there's nothing "very little" about the energy required to land on an asteroid let alone land on one, mine anything useful, and then return it to Earth. There's also the complexity of the vehicle needed to do the mining. Platinum (pure platinum mind you) goes for up to $37k per kg.

Even if we pretended that all of the technology existed and was donated and we just paid for a Falcon Heavy launch (about $150m to get about 17t of payload to an asteroid) we would need to ship back 4t of pure platinum to Earth just to break even. Even if you assumed that in the best case platinum was a thousand times more abundant in an asteroid than Earth's crust[1] that's still only 5ppm! You'd need your asteroid miner to process about eight million tons of ore (if my math is right) to get that 4t.

If our asteroid miner is 17t or so how long do you think it will take to process that much ore? The loan for the Falcon Heavy would have defaulted long before it would be even close to finished. There's just nothing asteroids have that ends up being worth the cost to mine them instead of mining them on Earth. Don't forget besides raw ore Earth also has lots of landfills literally filled with stuff than can be recycled and valuable materials extracted.

Materials in space are only really useful to an entirely space-based industry and the only way to bootstrap a space-based industry is to launch it from Earth which is basically just lighting money on fire.

[0] https://echo.jpl.nasa.gov/~lance/delta_v/delta_v.rendezvous.... [1] https://en.wikipedia.org/wiki/Abundance_of_elements_in_Earth...


Rocket fuel but now we are getting into circular logic...


the only sensible reason to mine space is to build in space.

Its not cost effective by and order of magnitude to mine in space and bring it back to earth.


No. :-)

Yeah, I know, not helpful. Hydrogen storage continues to be a huge issue. It is just so much more efficient to manufacture long chain molecules with hydrogen that you can then drive around in tanker trucks, store underground for pumping into units which burn them to recover the energy.

Once there is enough excess energy I expect you will see Fischer-Tropsch[1] type refineries that convert hydrogen and CO2 into liquid fuel rather than trying to ship around the hydrogen.

[1] https://en.wikipedia.org/wiki/Fischer%E2%80%93Tropsch_proces...


This should really just be the top post... I've been in this field for a while and totally agree. You're going to get vastly more practical applications out of liquid fuels or electricity stored in batteries.

The energy density of hydrogen is great on a mass basis but sucks on a volume basis unless it's compressed in which case the container is heavy enough that you lose all the gains. Plus compressed hydrogen is way harder to protect in an accident.

I think any theoretical application for hydrogen fuel cells you will find is using batteries in ten years. That's my opinion. The biggest advantage of a fuel cell is decoupling energy density from power density and flow cells solve that with batteries that use liquids that are easier to handle than hydrogen.


Methanol fuel cells are very inefficient at 30%. It is still better than internal combustion engine, but with H2 one gets over 70%. Still I agree that for niche applications like electrical cars in a very cold climate methanol fuel cells should be way more practical then hydrogen.


Bloom Energy's fuel cells are 53%+ efficient[1] on natural gas (Methane). Methane is easily liquified and transported and the infrastructure that supports this is widely deployed today.

[1] https://www.bloomenergy.com/datasheets/energy-server-es5-250...


Yeah, I've been keeping an eye on Bloom since they were still in stealth. It's neat and I believe their technology.

It seems to me like their fuel cell is designed to be a replacement for natural gas power plants and that it would better complement renewables than traditional fossil fuel use because it's probably extremely fast to respond to being turned up and down so it can follow unpredictable supplies. That's great!

Though to be fair, a combined cycle natural gas plant is also likely a lot more efficient than you are thinking, they're more like 50-60% than 30%. The gains may be very marginal... so maybe more for home use where if the electricity goes out maybe you still have gas (though in such an emergency I imagine you'd be more concerned with reliability than efficiency).

I can't really imagine these being used in cars, LNG does have infrastructure as you point out but if you think it's widespread I have to wonder where you live. In my life I've only ever seen LNG pumps in California and New Zealand. I think it heavily depends on what the country has kind of settled on as a standard fuel... but batteries are also 90% efficient at storing and releasing energy in temperate climates and I know solid oxide fuel cells tend to need to be super hot to get reasonable ionic conductivity don't they?

But also on the point of grid use, I should point out that the big difference between these fuel cells and a battery storage system is that these fuel cells as best I can see don't work in reverse. So you can't like take extra solar electricity and just run it backwards to make natural gas like you'd store power in a battery.

So I guess... within the error bars of the efficiency of a combined cycle natural gas plant and it can't store power? I hope it's cheap.


I designed an urban data center with them which was pretty cool. (fun fact, a lot of metro areas have way more gas capacity than they need, as opposed to electrical grid capacity) The bonus factor was not needing diesel generators onsite for backup power just keep a natural gas reservoir from the local supply.

We had some of these at Google powering the lab in one of the buildings with some standard Google type equipment racks. Easy start, rapid response, and N+1 or N+2 or 3 redundancy without needing a whole additional plant. (four of their 250kW units could cooperate to generate 1MW of power, and a fifth 250kW unit could provide surge and allowed for you to take a unit offline for maintenance without losing power.


That is very cool. I am surprised the cost came in under diesel generators, was it for noise/space/pollution more than cost? If they're legitimately cheaper for the same kW's as a diesel generator that's a different picture.

I still probably wouldn't use one as a portable generator but I'd definitely consider it as power backing for a hospital in a dense area where a big enough generator wasn't feasible (if it were in fact smaller).

Edit: I'm going to assume that if you pay for it you can get the kind of redundancy and automatic switching controls too with a diesel generator, I'm not actually positive they exist but I'd figure they would for the right price.


First, disclaimer, this was a while ago so the current data may be different (better or worse) but at the time, if you did the total cost of ownership thing, they were on a par with buying your power from natural gas 'peaker' plant and that was cheaper than a diesel generator. The diesel generator requires more maintenance than the Bloom boxes did. There are also the switchover circuity and battery bank that was required to hold up the data center between grid power loss and diesel startup. You can kill a modern data center by having a second power event before the standby batteries have enough charge to support another switch over. (which is why at least one company's data center stays on diesel until the batteries are recharged even if grid power returns).


Oh yeah, I suppose a lot more moving parts in a mechanical generator, I can see it ending up more reliable than a generator.

Seems unlikely to be more reliable than a battery backup system but I bet those battery sizes aren't nearly big enough (having lived quite a weird life and having actually researched industrial battery backup systems from GE too for some reason). I can definitely see it making sense.

Main downside I'd see is that if you're talking SF, I bet you the next earthquake takes out natural gas. If it's something important like a hospital that might be reason enough to have a big diesel generator since I know diesel fuel is so dense and you can ship it by truck if you have to. I don't really know how important your datacenter is in a disaster =)

I mean, honestly I mostly looked hard at Bloom before they left stealth mode but I kinda researched the same kind of stuff that they did in grad school and I'd be totally shocked if they were operating under 350C. Probably more like 500-600C.

If they had a major technical innovation I would imagine it's around decreasing the temperature that iron is catalytically active at but the solid electrolyte still needs to be hot enough to have good ionic conductivity.

My stuff from ages ago on this vague sort of thing: https://dspace.mit.edu/handle/1721.1/59228


Electricity has a big leg up here, we already have a great distribution system for this (with lots of capacity at off peak times) - battery technology (ie storage) has been the missing link here and it's rapidly being solved


Sounds like a commenter who knows what’s up.

Is storage a problem because of hydrogen loss due to diffusion, container embrittlement, or just the sheer nastiness of compressing and chilling something that can done blow up on you?


All of the above.


More like none of the above. The problems are long solved or greatly minimalized.


It's possible that they know this, but from a marketing perspective, hydrogen has a big ol' green circle/clean font plastered all over it.


Companies supplying equipment to do hydrogen electrolysis still matter in this equation though since hydrogen is feedstock in this equation.


Yes, but the number of them and their overall value is much smaller, maybe $100B.


At the scale we're looking at, it would still be around $2.5T. Actually more, since this process adds an additional inefficiency step.


So the future is internal combustion powered cars forever?


No, but the industry around tapping excess energy for use on demand is not moving hydrogen around. It is either putting into the grid (if user wants it immediately), into a battery if the user is attached to the grid and wants it later, or turned into a transportable hydrocarbon (probably methane) for storage and transport where electrical infrastructure is either too expensive or otherwise unavailable.


Except if you're making vast amounts of hydrocarbon such as methane at a cost effective rate, then there's little real incentive for any of the electric stuff. Internal combustion stays viable effectively forever.


Actually as the cost of electricity goes down the economics for electric vehicles stack. They are already a win from a maintenance perspective with fewer moving parts and fewer assemblies. As battery and regen technology improves wrt to charging, recovering charge from deceleration, and cost per kWH it just gets better still.

LNG is not as efficient as diesel in an ICE, electricity however is more efficient. Lots of discussion here[1]

[1] https://www.energy.gov/sites/prod/files/2014/03/f8/deer12_ka...


You can run the Fischer-Tropsch process and make any hydrocarbon fuel you want. Fuel efficiency of engines could exceed 50% in the near future. Remember, you're creating a scenario where we never figure out hydrogen, and therefore we'll be forced to make synfuels in vast quantities for a number of purposes. If you follow the nature tendency of capitalism, this will end give you hydrocarbon fuels so cheap that it wouldn't make sense to go with electric vehicles at all. At best, hybrids for that regenerative breaking ability.


A good friend of mine worked in compressor technology for 10 years. We had the fortune of working together for a few projects before he set off for Denmark to finish R&D of a new compressor for hydrogen fueling stations. He thought it'd only take 2 years. Took more like 5 years with all the certifications required and fine tuning parameters. So it's in use now in the EU.

When he was nearing the end of the project I asked if he thought the future of energy was battery technology or hydrogen or both? He answered both. The energy density of hydrogen is too good to ignore for commercial vehicles (buses, construction, 18 wheelers). The momentum in battery technology and infrastructure for (car) 'limited' range use is going to carry forward.

This friend now makes high end technology based liquor cabinets (chuckle). Probably too much pressure in his last gig.


I think your friend is making their judgement with a bit of myopia. EVs with batteries open a lot of opportunities that can't really work for any sort of liquid fuel vehicle, hydrogen included.

Batteries can be charged basically anywhere hooked to the electrical grid. They can also be charged in places with their own islanded power generation (renewable or not). So they can effectively be recharged anywhere they can stop for a while.

They can also be charged while in motion with electrified roads. A bus in a downtown area can be charged/run off overhead lines but then drive around outlying areas on their batteries. General utility overhead lines could also be used by any cargo vehicle with a pantograph for inner city driving.

Long haul EVs and construction vehicles are more likely to electrify as diesel-electric hybrids rather than pure battery or fuel cells. Their "recharge" profile is very different from passenger vehicles. Construction vehicles need to power their actual tools so they need super high density power they can really only get from diesel fuel.


Many construction vehicles use hydraulic motivation. The Diesel engine is only there to pressurize the hydraulic system. There’s no reason you can’t use an electric motor to motivate (The pumps inside) those vehicles, they don’t directly use the diesel output as they are built today.


You can use an electric motor to run the hydraulics but batteries are a poor way to power those motors. A diesel generator has far better energy density for a given volume. So I'm saying a diesel hybrid power train is better suited for construction vehicles than a straight battery electric.


I think he's wrong for some commercial vehicles even. Battery energy/power density hasn't plateaued yet and there hasn't been much in the way of vertical integration of raw metal input to cell output yet. Their input stock is an expensive already partially refined material (metal sulfates) rather than raw metal which increases the total production cost. Tabless battery cells will also increase the energy density.


I don't have any experience owning a concrete paving company [0], but there are many use cases where weight matters a lot (and refueling turnaround time also matters) like in aviation where a 45-60 minute ground turn is important to some carriers and where having the airplane get lighter as fuel burns off is also a feature from a performance, efficiency, and landing safety standpoint.

Without data to back this belief: I tend to think that synthesized jet fuel is probably going to be a technically easier solution than converting the jet fleet to hydrogen fuel (or to battery-electric storage). If you take yesterday's carbon from the atmosphere to make today's fuel, it's even carbon-neutral.

[0] - $NKLA's Director of Hydrogen Production/Infrastructure does. ;)


> I tend to think that synthesized jet fuel is probably going to be a technically easier solution than converting the jet fleet to hydrogen fuel

The US Navy has always been interested in that. They have surplus energy, after all, and incentives to manufacture their own fuel. It has not materialized it, for whatever reason.


We have known since WWII how to make fuel from scratch. Others have pointed the the FT process. Problem is when gas costs $2/gallon are you willing to pay $6/gallon? (when gas goes up your costs for FT tends to go up as well).

There are a few people who are - racers who need the custom blend of high energy (not to be confused with high octane though they may also run that) because that extra 1/10th second reduction in lap time matters.

FT might have a future in jet fuel where the weight advantages are worth the cost. I also believe that modern engineering could make it more efficient, but it will always be energy intensive to convert low energy materials into high energy fuels.


Probably because the raw materials to make JP-5/JP-8 are incredibly cheap and widely available currently.


The H2 cycle is quite wasteful. The touted "climate friendly" H2 will supposedly come from electrolysis. But electrolysis is a very inefficient process (as much as 65% of the energy is lost in the process), that's why industrial H2 is mostly produced from natural gas.

Then to transport H2 efficiently you need to compress it to several hundred bars. That process, again, consumes a lot of energy.

Then you'll have to transport and distribute that H2 around the country, to fill your car's tank.

All in all, the energy efficiency of the whole cycle is at most 15%, more or less the same as diesel.

H2 is a made-up response to the terrible ROI of wind turbines and solar when not backed-up with storage. That's how it's supposed to be the future. However when you do the math you quickly realize that it would be much more cost-effective to use nuclear power for combined thermolysis/electrolysis, that would get a way higher yield and ROI...


I work in renewables, the latest big buzz word is Power-to-X, meaning using renewable to generate hydrogen based fuels for things like jet engines and ships. I think it is 10-20 years away from becoming commercialised. Danish companies seem to be leading the way.


Would methane be an easier fuel to work with than hydrogen? Which is cheaper to produce from electricity? The wikipedia article[1] says methane could be produced from Hydrogen, but I believe there is a process to produce methane from water and CO2 directly.

Hydrogen is a feed stock for green steel production, which isn't commercially developed yet.

[1] https://en.wikipedia.org/wiki/Power-to-X


It really isn’t my area of expertise... all I see is that the investment firms that backed early offshore wind farms are now backing these Power to X technologies, as well as ‘energy islands’ which connect multiple offshore wind farms to multiple countries. Exciting times in renewables to be honest, it does feel like with enough renewable energy there is a path to a future without fossil fuels at all, we are just a long way from it right now.


I'm not an expert, I just follow this topic out of interest.

My general understanding is that Hydrogen doesn't make much sense for cars because it won't compete at that scale with the qualities (and size and weight and infrastructure) of Lithium Ion batteries, but where Hydrogen has a lot of room to grow/"disrupt" is in the long-tail of "heavy/occaisional diesel applications". This is things like ships (and cruise ships), festivals/food-trucks and other sorts of "off-the-grid" mini-grids, grid-scale batteries for load timeshifting (though Li-Ion looks quite competitive there too), possibly aircraft, and a very long tail of so many other ways that diesel especially gets burnt to generate mostly electricity (or refrigeration).


One of the qualities of batteries in cars is a place to store energy from regenerative braking. The best numbers for miles per charge are coming from the tech leader in the space (Tesla). The system of storing generated energy is important.


Absolutely, which is why Toyota's Mirai, the last living (why, Toyota, why?) attempt at a Hydrogen fuel cell car still carries the weight of Lithium Ion battery in addition to the fuel cell. I think at this point it seems pretty clear that if you have to cart around a Lithium Ion battery anyway you might as well go all in on an EV. I think it seems pretty clear that the Mirai is just a badly efficient EV with a strange battery mix and a very weird, very confusing, very expensive and hard to find "plug". I think that writing is well enough on the wall that everyone but Toyota seems to agree, and poor Toyota what are they even doing at this point, who knows.


I also think the Mirai is a dead end, but here are some things that might interest you:

* The energy density (kWh/kg) of the Mirai fuel cell system (stack, tank, etc) is probably 2x or more than its battery. Replacing it with a battery would probably cut range substantially.

* New car ramp-ups are weirder than I would've guessed, even when they're not pioneering a new fuel source and infrastructure. Eg the Prius took 8 years to break 50k sales per year, and it was a money-losing or breakeven proposition throughout that entire period. Later, of course, it became a breakout success story. [1]

* If fuel cells eventually matter (eg for heavy-duty applications), the Mirai could fail as a vehicle while succeeding as an R&D testbed.

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


> The energy density (kWh/kg) of the Mirai fuel cell system (stack, tank, etc) is probably 2x or more than its battery. Replacing it with a battery would probably cut range substantially.

I'm not sure how much that matters practically, however. The Toyota Mirai has an EPA estimated range of 312 miles and a base Tesla Model 3 is only a few miles shorter at 299 EPA estimated miles. They appear to be about the same size, though admittedly the Model 3 weighs 20X as much, if quickly searched numbers are correct. Consumers likely don't care about the weight difference other than how it impacts handling (and no one is complaining about the Model 3's handling) and indirectly felt impacts on road maintenance costs (and no one is paying attention to that at all, at least not right now in SUV/light truck-loving America).

(ETA: And it doesn't look like the weight scales the way that the size does in a comparable way to battery weight, density aside, and that Toyota can't just double/triple the tank size and blow EVs out of the water range-wise weight for weight or they would have already done so.)

> If fuel cells eventually matter (eg for heavy-duty applications), the Mirai could fail as a vehicle while succeeding as an R&D testbed.

That almost sounds like a valid reason for the Mirai to continue to exist. Certainly if you are a heavy-duty industrial user of Hydrogen such as a mining company using Hydrogen to replace diesel mining equipment, it might makes sense to have your support fleet of vehicles in/around your mines also capable of using the same Hydrogen infrastructure. That's looking likely to be a small corporate niche at best, so I'm still not sure that it makes sense in 2020 for Toyota to sell the Mirai as a consumer vehicle, but yes maybe calling it a gloom and doom for Toyota to be doing so is a bit strong.


The curb weight of the Toyota Mirai is 4,075lbs. The Model 3 is 3,552-4,100lbs depending on the configuration. As you noted, range is similar. What's very different is performance. The Mirai goes from 0-60mph in 9 seconds. The slowest Model 3 does it in 5.6 seconds. The fastest Model 3 gets to 60mph in 3.2 seconds. The Toyota's acceleration is limited by the power output of the fuel cell. Its maximum output is 114kW or 153hp. The Tesla's battery can output over 350kW, allowing it to drive much more powerful motors.


Read about the Hyundai Nexo and Honda Clarity. Far more companies are pursuing hydrogen than you think.


That's fair, I was unaware of the Nexo entirely and that the Clarity offered a Hydrogen version (in part because I'm among the vast majority of the US with no Hydrogen infrastructure and no attempts to bother selling it). That said, both Hyundai and Honda are clearly invested enough in full/proper EVs that these models especially feel like hedges/"compliance cars", versus Toyota that still hasn't seemed to commit to full EVs and still talks about the Mirai as if it were more than a hedge and was still the prime bet the company was making, which yes I think it an increasingly weird position in 2020. (Hyundai seems in a catbird position in some markets with the Kona EV right now, and Honda is still trying to figure out what company's electric future looks like, but clearly knows it is the future and the Honda e is incredibly cute and very "present", if unlikely to make it to the US.)

(Though a quick search just turns up that Toyota has definitely made an about face on that position in its Chinese operations last year and is finally planning an EV suite of cars with the Chinese market in mind. It's an interesting reminder that right now China is the EV market to beat in terms of how fast EVs have heated up in China. Interesting to watch if it will be the Chinese market that saves Toyota from US/EU mistakes like the Mirai.)


Hydrogen vehicles will not happen. The hydrogen industry is quite large today. The source of 99% of hydrogen? Fossil fuels.

If your goal is to get off fossil fuels, hydrogen is completely not the way to go.

Moreover, if it's performance that you want. Hydrogen 0-60 will be at best grandma speed.

Oh and let's not forget that it's hydrogen. In my jurisdiction, all pressure vessels for hydrogen must be inspected quarterly. Who wants to take their car in quarterly for inspections?


> Hydrogen 0-60 will be at best grandma speed.

Only if you don't have any batteries in the system, but in general I agree that hydrogen isn't going to be a major factor.


There are two separate questions here.

(1) Does hydrogen make sense as a store of energy?

(2) Will the various companies pushing hydrogen be financially successful in the medium term?

I think the answer to (1) is pretty clearly "no". But That doesn't strongly imply that the answer to (2) will also be "no". I can picture a world where hydrogen companies use hype to get funding, then use funding the get subsidies, then use subsidies to get more hype and install more infrastructure and generally push the concept to adoption. When big subsidies or externalities are on the table, it's very easy for the social/info/misinfo side of things to win and for the whole economy to do something stupid for 50 years. See, e.g., smoking, leaded petrol, nuclear power, etc.

As for why it doesn't make sense: hydrogen has essentially the same performance as batteries in terms of the car itself, but lower overall efficiency. So it costs you more energy, and it is less convenient because hydrogen requires a gas station whereas electricity doesn't.


In short, it makes sense where a gas station makes sense. Long cross country road trips. I can plug in my electric car at home, but when I'm half a day into a multi-day trip...


No, I don't think so. Nobody will get a hydrogen car just for trips.

People will get an EV and trade off the price of the battery versus the time it takes to fast-charge it.


Then you stop for a quick battery charge, stretch your legs and drive some more hours. No need to put on diapers to drive a long range EV with supercharging capability.


From a consumer's perspective, the thing I like about electricity and batteries is that it is relatively easy to be entirely self-sufficient. You can install solar panels and/or a wind turbine and be able to store their electricity in batteries without depending on anyone else. But you can't set up a hydrogen fuel cell factory in your back garden. That said, dependence on others is precisely what many incumbents in the industry want, so they may be pushing for such solutions for their benefit.


> But you can't set up a hydrogen fuel cell factory in your back garden.

Sure you can. Electrolysis is a middle school science experiment, and it's efficient to do at a small scale. In fact, it's often more efficient to do small scale electrolysis at fueling stations than to do it in a larger scale operation and ship it to the fueling stations.

(Splitting it from natural gas is cheaper, but then it's no longer clean energy, so...)


Well you can build a battery in your kitchen too with a potato, zinc nail and copper coin, but that's not going to be very practical for something like powering a car. What I meant was that, while you may be able to build a basic hydrogen fuel cell in your kitchen, if you want to safely generate, compress, cool, and store liquid hydrogen in the quantities required for something like a hydrogen fuel cell powered car, then you're going to need some expensive and bulky equipment, e.g. EUR60K-EUR90K (USD70K-USD105K) and 6-8m2 space for a system which just provides home heating not the additional piping etc. for car refuelling[0] (I'm not aware of any hydrogen production and storage facilities for car refilling which are currently commercially available for residential use, and wouldn't even be surprised if hydrogen refuelling systems ended up with some form of cryptographic protection mechanism to prevent refuelling from unsigned devices, e.g. home refuelling, for "safety reasons"). So for all intents and purposes, you're going to be dependent on a supplier for your fuel, which of course is what all the potential fuel supply companies want.

[0] https://www.homepowersolutions.de/en/product


A solar & battery setup is also going to cost 5 figures, and is going to take more than 6-8m2 of space. Both are well within the means of a small farmer or a remote northern community.

Hydrogen is stupid for a lot of other reasons, but dependence on central infrastructure isn't one of them IMO.


Practically nothing is absolutely self-sufficient... things break down, need parts


It's rarely mentioned how explosive hydrogen is and how this is supposed to be mitigated.

Another related problem has been solved though: Hydrogen cannot just be put in a normal tank, it dissipates through Hydrogen bonds which is a quantum mechanical effect.


Store it in solid-state form via Aluminum Hydride (Alane). Better stability & even less weight. Some fundamental scaling issues yet to be solved, but very much within reach.


How is it less weight? I'm assume this is counting the weight of the pressure vessel, and aluminum hydride reduces the pressure needed to contain the hydrogen?


Aluminum Hydride doesn't require a pressure vessel at all -- hydrogen is released at low pressure through heating -- which dramatically reduces the weight of the system.


This would be a lot more reassuring if the same (better storage in a bound form) hadn't been promised for at least 20 years, back then being solved in principle, but not being ready for prime time...


You would've said the very same about solar, wind, & grid-scale battery storage even 5-10 years ago. The commitment to scale is what made the "old" technology economically feasible. Incremental advances also played a role. This same equation is playing out for nuclear, Alane, and others. It's a matter of commitment & will -- both of which are gaining momentum.


What kind of tank can it be put in?

I honestly really appreciate gasoline for being such and incredibly portable source of energy. It gets a lot more hate than it deserves.


I don't hate gasoline/oil. I hate the stupid things we waste it on and the reckless constant expansion of that use. Our great grandchildren will curse us for wasting this liquid gold on making single use plastics and driving Disney cruise ships around and doing it all so intensely that we f'd up the climate.

We will always use it. Just hopefully more judiciously?


> It's rarely mentioned how explosive hydrogen is and how this is supposed to be mitigated.

This is left as an exercise to engineers :)

But as a matter of public perception, what about the invisible flame? All it takes is a bunch of horror stories.


Could you expand on;

> Hydrogen cannot just be put in a normal tank, it dissipates through Hydrogen bonds which is a quantum mechanical effect.

This goes against my understanding of a hydrogen bond.


Yeah you're right, it's not Hydrogen bonds, it's called "Quantum Tunnelling". Basically it means that an object can pass a potential barrier with a certain probability even if it doesn't have the energy to do so. The effect depends of course the height and width of the barrier, the temperature but also on the mass of the object. (Therefore it's mostly relevant for Hydrogen)


At least for fueling airplanes, hydrogen doesn't appear to be non-sense. I initially suspected that Airbus was just putting out some hot air, but now I think there's a 50-50 chance they are serious.

The basic idea is: hydrogen is by far the chemical fuel with the highest energy density in terms of mass: about 3 times the energy density of gasoline or jet fuel. However, liquid hydrogen has very low (mass) density so in the end the volumetric density of hydrogen is about 3 times lower than that of gasoline. But that works out ok. For example, an Airbus A320 has a max takeoff weight of about 75 tons, out of which about one third is fuel. If you replace that fuel with the liquid hydrogen that has the same caloric content, you need only a third of that in terms of mass (so about 8 tons), but since that occupies pound for pound 3 times as much volume, these 8 tons will fill the same volume as the 25 times of the regular jet fuel. This is great news: you don't need to modify the shape of the airplane, you don't need to make it like a beluga to fit bigger tanks.

Moreover, there are reasons to believe hydrogen-fueled engines have higher efficiency, maybe 10-20% higher. On top of that, if you reduce the takeoff weight from 75 tons (which is 25 tons of fuel + 50 non-fuel) to only 58 tons, you end up needing less fuel to carry this airplane around. It's the curse of the rocket equation working in reverse, in other words it's a blessing: using a more dense fuel means you need to carry less fuel, therefore you need to carry less fuel that moves the fuel, etc. In the end you probably get a few percentages of savings right there for free.

So, these are the pros.

What are the cons? You need cryogenic storage, you probably need tanks with a special interior lining to limit the leakage, you need infrastructure to move the hydrogen around. On top of that, there's a chance you increase the NO2 pollution (and it's not like an airplane can carry with it a catalytic convertor to take care of that).

An obvious pro is no CO2 emissions. An obvious con is the cost of producing the H2.

All in all, I think there's a fair chance that H2-powered airplanes could be a good idea.


We have to use energy from another source to create hydrogen -- so its only use is as an energy store, not as a source. Hydrogen is competition with batteries, not with other energy sources.


> Once we have better batteries for storage, what problem do we have left that requires hydrogen to solve?

An obvious short-term problem is weight-efficiency - hydrogen is still far better than best batteries, and weight is of crucial importance on aircrafts, so if you want to fly green - jets, turboprops, electric propellers - you can use hydrogen.


Ok, let's see here...

Hydrogen has 142 MJ/kg and at 700 bar is 42 kg/m^3. "At this pressure, 5 kg of hydrogen can be stored in a 125-liter tank." (https://energies.airliquide.com/resources-planet-hydrogen/ho...). That's 710 MJ in 125-liter tank.

I couldn't find the stats on a 127-liter tank, but a 49-liter, 300 ft^3 tank is 139 lbs, or 63 kg. That would hold about 2kg of hydrogen. (https://www.mathesongas.com/industrialgas/pdfs/Industrial-Cy...)

That's 280 MJ in 65 kg, or 4.3 MJ/kg. A lithium ion battery is 0.36–0.875 MJ/kg, but jet fuel is 43 MJ/kg. (https://en.wikipedia.org/wiki/Energy_density)


I dont really know the correct terminology, but I think batteries also have an issue with "burst" power loads.

Think like the difference between a battery and capacitor. Some alternatives like super capacitors may bring greater power, albeit not at the same total storage?

Is this correct understanding?


Power density for supercapacitors is higher than batteries, but there aren't many applications that need more power than what the best lithium polymer cells that provide. 100C cells are readily available, which can discharge their full capacity about 1/100 hour ~= 36 seconds. These are the cells that power high-speed RC planes and handheld power banks that can jumpstart your car. One caveat with high discharge rate batteries is that cycle life is not necessarily any better, and charge rates are usually much lower than discharge rates, so sometimes you will see supercapacitors used to temporarily hold a large amount of energy.

Hydrogen has high power density in the sense that you can let it out of the tank very quickly and ignite it, but unless you want an explosion you are still constrained by an ICE motor or fuel cell. Those are much less power dense than batteries.


Yes, it is correct concerning burst energy from capacitors. The #2 heavy equipment manufacturer has a large capacitor in their regenerative excavator (which sees the most sales in Europe, not North America).


That depends on the battery chemistry and construction. Li-Ion and flooded lead-acid batteries can produce fairly high surge currents. Coin-cells on the other extreme cannot.


Even with relatively high rates of discharge, my non-expert understanding is that they are still far slower than capacitors. E.g. if you're trying to build a railgun or absorb power from lightning strikes you better be using capacitors (which might be charged from a battery).

That last part of sort of the point, while batteries and capacitors serve the same theoretical function (storing energy), in practice they are used differently. One as a sort of buffer, the other for actual storage.


While true, even a battery pack used for small RC cars can output several hundred amps in bursts or even constant discharge.

Take this[1] pack of four cells for example which is rated for 450A constant discharge and 650A bursts. And there are more extreme packs like this[2], still four cells, which is rated for over 1100A burst discharge.

Sure a capacitor might do better for a brief instant, but li-io batteries can put out some serious energy quickly.

[1]: https://hobbyking.com/en_us/graphene-5000mah-2s2p-hardcase-w...

[2]: https://hobbyking.com/en_us/turnigy-rapid-8000mah-2s2p-140c-...


"Once we have better batteries for storage, what problem do we have left that requires hydrogen to solve?"

Well... also storage. The cheaper it becomes to store energy as hydrogen the better. The fact that the overall process is less efficient than battery storage does not mean hydrogen will not be used.

The amount of energy that can be stored in batteries is relatively small, and there are only so many viable hydroelectric storage sites. We'll need more than one way to store energy on the grid level.

Hydrogen also enables countries to export energy. For instance, quite a few oil-producing countries receive lots of sunshine. Moving away from oil, they could produce hydrogen using solar power and export it elsewhere in the world. That might be less efficient than transporting the electricity directly, but much easier politically: just imagine building a power line from Saudi Arabia to Germany. How many countries would have to collaborate to make it work?

Edit: hydrogen can also replace other substances in process engineering. For instance, there are ways to produce steel with hydrogen instead of carbon-based fuels as a reducing agent.


Hydrogen is not great as a means of exporting energy, it is, unfortunately, hard to transport. Hydrogen is about 1/3 the energy/volume density of natural gas - so much worse than liquid fuels - and has a very low boiling point so liquified hydrogen is very (energy and cost) expensive to produce.

Pipelines can work though, if you can get them built. The lower energy density vs. natural gas is offset by a higher flow speed. And the energy transported by one pipeline could be much greater than a HVDC power line.


"Hydrogen also enables countries to export energy."

Expanding on that... Bitcoin also enables countries to export energy.


Absolutely not for normal cars, that battle is long lost. Its baffling that some companies still pursue it. Often with government funds of course.

For heavy transportation hydrogen might have a small market for extremely long distance transport. But if you ask me, even that market is practically out of reach already. Companies like Daimler want to have a hydrogen truck 'in the second half of the decade' but by then batteries should clearly outclass it. And that is before you have even solved the infrastructure issue.

I think equally in storage, it makes little sense to use hydrogen outside of some specialized applications. There are other chemical alternatives as well but here I would still bet on the battery for a much, much, much larger part of the grid, both home and grid-scale.

The two industries where hydrogen might be interesting is long distance container ships and flight. There it still has some chance against batteries but for the majority of flight, battery is within clear distance of known technology. There is long range flights, but for those hydrogen competes with other chemical fuels as well.

I don't know enough about the shipping industry. You can do things like using dimethyl ether but you essentially need hydrogen to produce that, so I guess it sort of counts. The same goes for things like methanol.

The predictions of 'Bank of America' seem like total nonsense to me. The battery is riding a cost curve with governments and privates from all over the world putting massive resources into it. Fuel cells have some government that love it, and some companies that believe in it but its not near the same scale of investment.

Lots of nonsense has gone up if it is connected to something clean energy, its the 'next Tesla' hunt.


It's uncertain.

Despite what people say, main problem with hydrogen (when compared to other energy storage/sources) is not efficiency or safety, but overall cost. What drives the energy market is $/watt for a given solution, not raw efficiencies.

That said, cheap hydrogen production is an active research area. Some ideas under research include hydrogen producing bacteria and analogs to photosynthesis. Some methods, even appear to be nearly perfectly efficient [1]. Now could hydrogen become a major source of power in 10 years? Probably not, as there are lots of costly problems to overcome - not just in production. Will it be in the distant future? Probably, given that it's the most abundant element in the universe, the power source of most stars, the easiest element to do fusion with - that it's hard to fathom us becoming a successful space faring civilization without it...

https://www.eurekalert.org/pub_releases/2020-06/su-shp060220... [1]


Hydrogen is not source of power unless you are doing hydrogen-related research and this gives you power in the organization you work for.

Power is rate of energy transfer. https://en.wikipedia.org/wiki/Power_(physics)

You would not typically call something "source of power". Rather, we would typically discuss "source of energy" which is then converted to power by expending it (transferring/converting) at different rates depending on application (think in terms of your power at your academia letting you force undergraduates doing work for you).

As to sources of energy, the only way hydrogen can be used as a source of energy is through fusion which is currently unattainable on commercial scale in 10 years. There are also no sources of hydrogen that can be mined or otherwise exploited on Earth.

There exist, arguably, fossil fuels that are rich in hydrogen. A cell using such a fuel could technically be called hydrogen fuel but in normal use they are typically called cars.


10 years? No.

Batteries are killing hydrogen on scale and cost improvement. We may already have reached a "curve superiority" by batteries much like DRAM vs MRAM, as in MRAM may have had overall potential superiority to DRAM, but was way too late to the party to beat out DRAM's constant incremental improvements. The lead on economies of scale mean and allocation of research means that hydrogen cannot "catch up".

Even if massive investment in hydrogen started, presumably in transportation, it is probably 5-8 years from reaching the point Tesla was with the release of the first Model S.

Which cedes Tesla and battery tech another 5-8 years of "main curve" improvement. Consider that the great leap of lithium metal/solid state seems an almost guarantee in 10 years time, and likely 5 years.


You got to look at both efficiency and storage costs. Batteries are really expensive but efficient.

For short term storage batteries make more sense because short term efficiency affect economics more.

But when you store long term that storage needs to be cheap. Hydrogen or a synthetic fuel derived from hydrogen gives you that.

There are also use cases where batteries are too heavy. For long flights batteries are too heavy but hydrogen will work. After all plenty of rockets use hydrogen fuel, none use lithium-ion.

Then there are plenty of industrial process which can use hydrogen. E.g. a lot of coal is used as a reducing agent when producing silicon, iron etc. This cause emissions when creating solar cells e.g.

By using hydrogen as a reducing agent instead we don’t get CO2 emissions.


I have no clue what will happen in the next 10 years.

Hydrogen Fuel Cells seem like a marginal solution - it can make sense in some situations. For example the military loves it because they can make electricity without the noise of a generator.

Fuel Cells could make sense for some transportation problems, but they never will for other problems.

H2 gas will always have problems with pressure, containment, embrittlement, and conversion/production costs.

However, be on the lookout for other storage mediums, some of which are mentioned by other comments - Aluminum Hydride (Alane), Methane, Ammonia.

In fact, look out for anything that can take excess electricity from renewables and store it. Aluminum Air batteries for instance.


I really doubt it. Hydrogen is explosive and extremely hard to contain. It also burns too hot and too fast for regular engines. It's basically a bad fuel all around. If it was any good we would be using it for heat instead of natural gas.

If we're producing chemical fuel on a large scale, hydrocarbons are a nearly perfect form.

More likely, some kind of battery breakthrough will replace lithium ion. We only need 3-5x current power density to replace fuel for most uses. Lithium ion was about 3x better than NiCad when it came out. If we can repeat that just once more, there will be a battery revolution


EVs and different chemistries based energy storage (Zinc air, LiFePO4, liquid metal batteries) will win first and hydrogen will help with curtailment issues of excess renewable energy for longer duration storage.


> It seems we are getting to the point where wind power and particularly solar power are now cost effective.

Hydrogen is made from fossil fuels, so I don't understand this point.

Yes you can make hydrogen from electricity, but why. There are better uses for excess electricity.

> Ask HN: Is hydrogen likely to be a major source of power in the next 10 years?

Nothing will happen in 10 years. But it might be clear it'll be the way forward in 10 years.

Batteries are not a solution to hydrogen generally.

The answer to hydrogen I think will be the same as now, fossil fuels. But we will work on cleaning the PM 2.5 coming out. That will be big in 10 years.


Straight forward answer is energy density needed for shipping and flight. Even if we get super-duper mega battery with 100%+ more kWh/kg in 5-10 years it will take time to scale it up while hydrogen delivered through synt. fuels is here. It seems that the Danish are betting on it: https://www.energyglobal.com/special-reports/28082020/haldor...


I am surprised to see only three comments mentioning hydrogen embrittlement here. It's the central issue with H2, and I know of nothing that will overcome that.

Also, batteries are not going to be unseated because they have a massive head start here. The vast majority of the industrial R&D effort either is (or will be shortly) behind them now. We've seen this movie before: LCDs vs. plasma displays, silicon vs. gallium arsenide, RCA's Capacitance Electronic Disc (CED) vs the optical LaserDisc.


Because it's not an issue anymore. It's only a problem with certain types of metal alloys, which can be averted using different alloys, plastic linings, and using carbon fiber where possible.


Show me that link, and I'll happily take a look!


It's not that there's a big debunking article, but rather people have used hydrogen for decades and not encountered those problems. For instance, hydrogen pipelines existed since the 1930s, showing the problem is eminently solvable: https://en.wikipedia.org/wiki/Hydrogen_pipeline_transport


From the article you linked to:

"For process metal piping at pressures up to 7,000 psi (48 MPa), high-purity stainless steel piping with a maximum hardness of 80 HRB is preferred.

Composite pipes are assessed like:

carbon fiber structure with fiberglass overlay perfluoroalkoxy (PFA, MFA) polytetrafluoroethylene (PTFE) fluorinated ethylene propylene (FEP) carbon-fiber-reinforced polymers (FRP)"

All of these are expensive options compared to battery electric technology, which leverages widespread infrastructure (think electric power lines!), as well as rapidly decling costs (see Tesla's Battery Day announcement on 9/22/2020).


They're not. Some of these are just cheap plastics. Just line a steel pipe with these chemicals and you're good.


One problem with hydrogen is storing, it is rather lightweight, even when compressed to be liquid, which is good, but it is hard to contain in a steel tank. Source:

   Few materials are suitable for tanks as hydrogen being a small molecule tends to diffuse through many liner materials and hydrogen embrittlement causes weakening in some types of metal containers 

  taken from https://en.wikipedia.org/wiki/Hydrogen_economy


Elon on hydrogen: https://www.youtube.com/watch?v=yFPnT-DCBVs


Could you link to the research report of Bank of America?


Hydrogen could be a major source of power in the late stages of the energy transition. Batteries are good for storing power for hours. Hydrogen can be stored for months, so can mitigate seasonal cycles and severe weather conditions.

But there's little point investing in long-term energy storage while there remains market opportunity in short term storage, which will still be the case in ten years.


No. Liquid H2 is incredibly fluffy. You get more H's (and crackable bonds) in hydrocarbons, as elsewhere discussed.


See also: https://www.forbes.com/sites/arielcohen/2020/09/23/airbus-un...


It won't be a "major energy source", batteries are not even going to be a "major energy source" in most regions and they've been under more development for much longer than fuel cells. Fuel cells will have a niche role, but not a major role, for at least a decade.


I recently saw an article saying that elements of the airline industry have tested batteries, and have decided to go with hydrogen ... on something like your 10-year timescale. As I recall, the maximum range of hydrogen flights would be on the order of half that of jet fuel.


We have become used to a world where hydrocarbons are the energy source, certainly for portable energy. Historically it has not been the case that there is one predominant source of energy. I think that we will go back to that state of affairs.

So, yes, hydrogen powered fuel cells are likely to be a major source of power in the next 10 years. It is just that there will be other major sources of power as well. Different energy storage systems have different attributes and are viable in different applications.


as per the other comments -- not a source of energy but a store of energy

whether the energy stored is green or not


Imo the only world in which hydrogen becomes a viable energy currency is one where there's vast amounts of highly centralized power generation, e.g. fusion (but possibly also off-planet generation) that needs to be effectively distributed to agents that can't be connected to the grid, like planes or to a lesser degree cars.


Only if by hydrogen, you mean tritium.


Link to the Hydrogen Primer?


(Shameless promotion: if this stuff interests you, check out my brand-new course https://www.terra.do/hydrogen-economy)

There is a developing consensus that in order to get to "net zero" by 2050, hydrogen will need to be a significant part of the energy mix. For example, here [1] are two forecasts that predict H2 will account for ~10% of global energy use. This could just be dismissed as "hype" but across China, Europe, and elsewhere, real money is going into electrolyzers and fuel cells (at, I dunno, 500% YOY growth?) which I think is why stocks are acting as they are.

Here's a list of where H2 fits in, and other sustainable alternatives. (Of course, for all entries, "indefinite fossil fuel usage" is an option in some sense!)

Current H2 usage. Fossil-derived hydrogen is a cornerstone of the economy (for eg fertilizer), and responsible for a few % of global GHG emissions. Replacing that with low-carbon-intensity hydrogen would be a major win by itself. Kind of by definition, there are few alternatives here.

Other industrial uses. The production of steel, plastics, etc are complicated systems, each too complicated to explain here, but eg HYBRIT [2], a fossil-free steel plant, could reduce emissions from Nordic countries by 7-10%. I have not heard of many alternative decarbonization pathways here.

Non-passenger-vehicle transport (trucks, ships, planes, trains). Molecular fuels have higher energy density than current-gen batteries by 1-2 orders of magnitude. I predict that these vehicles will mostly not be battery-powered. Biofuels are another prospect. "E-fuels" (sourced from H2 and an atmospheric carbon source) are another.

Lots more. Intermittent renewable valorization, building heat, power storage, gas pipeline admixtures... but this comment is already too long. I also like this story [3] about the Intermountain Power Plant, the largest H2 project in the US that I'm aware of.

[1] ETC: Page 38-39 of https://www.energy-transitions.org/wp-content/uploads/2020/0...

BP: https://www.bp.com/en/global/corporate/energy-economics/ener... https://www.bp.com/en/global/corporate/energy-economics/ener...

[2] https://www.reuters.com/article/us-sweden-steel-hydrogen/swe...

[3] https://www.latimes.com/environment/story/2019-12-10/los-ang...


Hydrogen has started to become more prominent since (most of) the world started talking about getting to net zero carbon emissions, rather than, say, 80% cuts by 2050. A net zero system needs to meet several needs which renewables and nuclear don’t meet well and which batteries don’t yet do much to help with.

The three biggest problems that come to mind are (a) meeting peak energy demand, (b) process heat for some industrial processes where there’s no current adjective to a burner, and (c) fuelling long distance vehicles like planes, ships and long distance truck and train routes away from big electricity infrastructure.

Hydrogen looks like a good candidate to be at least a part of the solution to all of these as it is currently expected to be the cheapest chemical fuel to produce at sufficient scale in a net zero manner. Multiple (proprietary, not all publicly available, but see [0]) predictions say that as scale grows hydrogen from electrolysis powered by dedicated renewable plant could reach prices comparable to those of natural gas in most parts of the world (not the US) today. The recurved advantage of hydrogen could disappear, but there’s no clear alternative today.

For (a) meeting demand peaks - a big problem in cold regions where space heat is a big part of energy demand and the occasional winter is very cold - hydrogen can be produced when the wind blows and the sun shines and then stored. This is easier in places with the right geography to store hydrogen directly (e.g. salt caverns), but hydrogen may still play a part elsewhere. Batteries are energy efficient, but very cost inefficient at long duration (in this case, months to years) energy storage.

For (b) fuelling industrial processes, hydrogen can usually be used as a direct alternative to oil or natural gas. Sometimes this would require new plant, sometimes not.

For (c) transport, batteries are more likely to develop to solve many challenges. Where they don’t (my guess: all but short distance aviation and shipping; very long distance trucking), hydrogen is the most likely chemical starting point to create net zero fuels - maybe hydrogen itself, maybe synthetic hydrocarbons, maybe ammonia.

Note that none of these is an immediate challenge anywhere today, so hydrogen is seldom a commercial solution. But they all will be to get a net zero future.

Now none of this means today’s hydrogen companies will be successful. In fact, based on past transformations, most will probably fail. And those that do succeed and grow, whilst technology driven companies, may well not be anything like either software developers or oil producers in terms of return. I’d guess a mix of smaller technology providers (maybe c.f. Arm if they’re very lucky) and something more like regulated utilities or merchant renewable developers: relatively low risk and thus return, capital intensive industries.

[0] https://about.bnef.com/blog/hydrogen-economy-offers-promisin...


This is a really good breakdown of the current state of affairs, thank you!


Last time I checked, hydrogen looked pretty bad. Low energy density and you can't draw a lot of power from it. Hydrogen initiatives always seemed to come from governments, which is usually a sign that they're not a viable tech.


No, unfortunately not.

People seem to misunderstand how institutionalized most of our infra is.

If we had a 'great car, and great tech' today in 2020, it would take 10 years to start to see common adoption.

But we really don't have proper hydrogen solutions and infra taking shape, so it's unlikely anything material will happen in 10 years.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: