9 ms·

My unpopular opinion is that H2 is an overall very poor solution to our current problems. It replies poorly to the electricity storage problem, because the prod

by wazoox 6y ago

My unpopular opinion is that H2 is an overall very poor solution to our current problems. It replies poorly to the electricity storage problem, because the production chain efficiency is really low, as low as 15-20% : electrolysis has a ~45% efficiency, then compressing the H2 to 700 bars consumes 30% of the stored energy, then you must add distribution and transport (and mostly can't use the huge, existing gas transportation infrastructure). While batteries have an efficiency over 90%, ditto (already existing) HT network. Also H2 isn't great as a dense power storage for vehicles either; the H2 plane is but a pipe dream, because it requires entirely new ways to store energy aboard, therefore probably requires entirely new planes with new architectures (like flying wings), huge new airports infrastructures... All in all, methane would be a saner, simpler target : you can create methane from water, CO2 and electricity; we already have all the infrastructure to transport, store and distribute methane; and using methane in vehicles of any kind is a no-brainer and requires no breakthrough or radical change. I don't believe a word of all the H2 hype.

usrusr 6y ago
> then compressing the H2 to 700 bars consumes 30% of the stored energy, then you must add distribution and transport (and mostly can't use the huge, existing gas transportation infrastructure). I recall reading somewhere that (in some/many countries) existing gas infrastructure is already specified to tolerate a meaningful (if small) fraction of H2 in the mix and that this fraction isn't currently maxed out, implying that it could be used as a sink for a quite sizable amount of new H2 production capacity as is. Background: historically the network was designed to tolerate a much higher H2 fraction because the initial gas source was a conversion process from coal that happened to yield a H2 fraction considerably higher than what is currently distributed. So as long as conventional gas use continues you'd have zero storage problem for new H2 production until the existing tolerance limit is maxed out. Only switching to anything close to 100% H2 would be a storage/distribution problem. Bonus implication: retooling existing infrastructure back to tolerate that old, higher H2 fraction would be considerably (orders of magnitude) easier than building a network for pure H2. I still share your main view, that there tends to be too much hype for H2. E.g. be wary of people who use an imminent "H2 future" as an argument to abandon all other projects. But that doesn't negate the value of real-life H2 projects at all: chances are we won't satisfy easy H2 use cases any time soon and the things that would make difficult/impossible H2 use cases difficult/impossible shouldn't be used as arguments against production until the easy use cases appear in reach to getting saturated.
- wazoox 6y ago
  I don't know for other countries, but in France Enedis (in charge of the distribution network) indicated that the network can support no more than 6% of H2. IIRC higher proportions of H2 would actively deteriorate tubes and stuff (I don't know the details, but apparently it can even react with certain materials and make them brittle for instance).
  - rjsw 6y ago
    It reacts with fairly common materials [1], steel is the main one. [1] https://en.wikipedia.org/wiki/Hydrogen_embrittlement https://en.wikipedia.org/wiki/Hydrogen_embrittlement
  - usrusr 6y ago
    That's roughly the number I was talking about. Subtract current H2 percentage in the system and multiply the result with network throughout (plus storage headroom). That's quite a lot slack to pick up before running out of storage for regenerative H2 production. Sure, those 6% are a local maximum which further limits how much a hypothetical production point could push into the network, but that'd be more of a tactical problem (where to deploy the tech) than a universal limitation. And a solvable one, because the inputs to the conversion are quite transferable.
  - hokkos 6y ago
    Enedis is in charge of electricity distribution, GRDF is in charge of fossil gas distribution, and also one of the partners of the grhyd experiment to inject H2 in fossil gas networks, and they went to 20% ! http://grhyd.fr/ http://grhyd.fr/
funcDropShadow 6y ago
And methane has the added benefit of being able to be used in unchanged home heating systems, at least in large parts of Germany and probably of Europe. And there is an existing distribution network in the form of pipelines to most houses. And there are large storage facilities, like the strategic reserves for natural gas. If you factor in the saved CO2 due to the fact that you don't have to change most home heating systems. And that you can easily retrofit lots of gasoline engines in already existing cars. Then I don't get why Germany started to support EV at all.
- manfredo 6y ago
  The problem is that synthetic methane requires a source of carbon. Extracting the fraction of a percent concentration of carbon in the atmosphere is not feasible. Hydrogen is one of the inputs for the sabatier process, so if the lower energy density by unit of volume isn't a concern it makes more sense to just use the hydrogen directly. The reality, though, is that none of the proposed energy storage solutions are easier than just building nuclear power plants. France already generates most of it's electricity from nuclear power. Belgium recently surpassed 50% nuclear power generation. It's more feasible to just avoid the unsolved energy storage problem altogether.
  - epistasis 6y ago
    Some startups are working in direct air capture of CO2. I've linked one in another comment here, Carbon Engineering: https://en.wikipedia.org/wiki/Carbon_Engineering https://en.wikipedia.org/wiki/Carbon_Engineering They think that at scale, a ton of CO2 will cost $100 and take 5-6GJ if energy. Totally agreed that it is far cheaper to just never emit in the first place. But as far as building new nuclear, France has not been able to succeed at this economically, and similarly the UK's plans at building new nuclear are coming apart completely. And even if we are able to decarbonize, with whatever technology, be it nuclear or not, we will still need to get to negative CO2 emissions in the coming decade. While some of that will take the form of CO2 sequestration in the biosphere, such sequestration is only temporary, and we will eventually want geologic sequestration. So we must solve the capture problem by some means. Perhaps it will be with something like Carbon Engineering's process, or maybe something else.
marsokod 6y ago
Regarding the planes, the Soviets had a plane working with hydrogen (and methane I believe). It's not particularly impossible, and at least much more doable than using batteries (for commercial jets). The flying wing is a design improving the efficiency know for a long time (Airbus guys teaching us about it during my studies 15 years ago), but creates additional difficulties, on top of which the impossibility to evacuate the plane fast enough. You would have the same benefits whether your plane is using hydrogen or kerosene. I still do not see planes using hydrogen. Just doing the math on how much electricity would be needed makes it really unlikely.
- wazoox 6y ago
  In the 70s engineers in CEA in France modified some cars on the campus to run on liquid hydrogen. Apparently burning hydrogen in a combustion engine is pretty straightforward. I think that's the plan about future H2 planes, make them use turbofans slightly modify to burn H2 instead of kerosen. However the big problem is storing H2 onboard. Most of the fuel is in the wings, and very cold liquid hydrogen (icing) as well as very heavy high-pressure tanks won't fit there. That's why I mentioned that an entirely new concept may be necessary to make room for the fuel tanks inside the fuselage.
  - rjsw 6y ago
    The Airbus proposals from a couple of weeks ago had the fuel tanks in the rear of the fuselage.
    - wazoox 6y ago
      Yes, that's the problem. You embark much less energy as fuel, lose lots of room for passengers and cargo; I very much doubt this could make sense economically. My brother-in-law who is an aeronautical engineer at Airbus isn't convinced, either.
the8472 6y ago
> then compressing the H2 to 700 bars consumes 30% of the stored energy, then you must add distribution and transport You don't need any of that for grid storage. You can pump it down into a salt cavern in gaseous form (previously used for natural gas storage) and pump it back for on-site power generation when the grid needs it.
- pfdietz 6y ago
  Also, the energy of compression can be substantially recovered when the hydrogen is used.
pfdietz 6y ago
Your objection misses the mark. There is no singular "energy storage problem". There are various energy storage problems. All hydrogen (or batteries, or other storage technologies) have to do to be viable is address one or more of them. No storage technology can solve all of then, and no single storage technology has to. Hydrogen is best suited for long term and infrequently tapped energy storage. For that use case, it is vastly superior to batteries, even if its efficiency is much lower. And this use case has to be solved to get to a 100% renewable grid. Even aside from that, realize that today 6% of all natural gas consumed in the world goes into making hydrogen, mostly for manufacture of ammonia. So this use ALONE -- which amounts to 700 cubic kilometers of H2 (at STP) per year -- is a vast market for renewable hydrogen production.
- jiggawatts 6y ago
  You miss his point. Hydrogen is the worst at everything, if compared to all available alternatives, not batteries only. Other technologies outperform it very substantially in almost every metric, often simultaneously. Hydrogen is energy inefficient to produce, space inefficient to store, dangerously explosive, reactive with common engineering metals, and unproven. The one thing hydrogen is optimal for is mass efficiency, but methane is almost as good and far easier and safer to store than hydrogen, and has a much higher energy density by volume, which matters more than you'd think, even for planes and rockets. The new super-heavy rocket designed by Space X uses methane, and the current ones use kerosene. For grid storage, the sole advantage of Hydrogen -- its low density -- is utterly pointless. Nobody cares how heavy building-sized storage is. Efficiency matters, and the electricity-to-hydrogen-to-electricity round trip efficiency is woeful. There's no clear way to make this anywhere near as efficient as batteries either. Not to mention that the huge benefit of batteries is nearly instant (single-digit millisecond!) reaction time, allowing power demand fluctuations to be efficiently smoothed out. Hydrogen can't do that, even with fuel cells. Lastly, we make hydrogen for ammonia to use as fertiliser. This doesn't in any way mean that it this is efficient or worthwhile for power storage.
  - pfdietz 6y ago
    > Hydrogen is the worst at everything, if compared to all available alternatives, not batteries only What's better for seasonal storage?
rstuart4133 6y ago
> My unpopular opinion is that H2 is an overall very poor solution to our current problems. I think you opinion is based on dated ideas. The interest in hydrogen is not being driven by electric cars, or as a battery replacement. It's being driven by one thing: if the price of renewable energy keeps dropping, it will be cheaper to make hydrogen than frac gas or mine coal. Even better, one accepted way to deal with renewables intermittency is to over provision. Over provisioning means we produce lots of unwanted electricity that is effectively free. If you can bottle what get get for free and sell it: profit. To put this another way, the hydrogen will be used to produce heat. While lights, cars, planes and other things get all the attention, plain simple heating things is the number one user of energy. We use it to cook food, warm our houses, take showers, wash our plates and pants, iron our clothes, melt silicon ingots, make steel and concrete. Just about everything we use need heat in its production. As you say storing hydrogen is problematic. As possible solution of storing it as anhydrous ammonia has been thwarted by our inability to reverse the reaction and get the hydrogen back. That little sticking point aside, it's nearly ideal. We efficiently produce enormous quantities of it now for fertilizer, and have decades of experience in ensuring your average job farmer handles this rather obnoxious substance safely. But, where there is a lot of money to be made, people have an incentive to go looking. And in Australia at least it dawned there was a _lot_ of money to be made. Australia is the 5th largest exporter of natural gas in the world. Like ammonia natural gas is also difficult to handle. To transport methane efficiently it has to be a liquid, which is done by cooling it to -26O F and compressing 4psi. While exports are great now, it's on the nose. China, Australia's biggest customer, plans to be carbon neutral by 2060. But looming disaster could be circumvented as Australia has huge reserves of sunshine at it is mostly desert, it is politically stable and is very good at things requiring large coordinated infrastructure investment like food production and mining. Hydrogen would be a natural fit for us, if we could solve the storage problem. So, they did. Or more precisely the scientific arm of the Australian government, CSIRO, did. They came us with a metal membrane that converts the ammonia back to hydrogen. https://www.csiro.au/en/Research/EF/Areas/Renewable-and-low-emission-tech/Hydrogen/Hydrogen-membrane https://www.csiro.au/en/Research/EF/Areas/Renewable-and-low-... I guess the plan is instead of taking delivery of a tank of methane, you take delivery of a tank of ammonia, and you convert it back to hydrogen and burn it. Burning hydrogen sounds so low tech, so inefficient. It is low tech, but burning it for heat production is 100% efficient. And remember the driving force behind this is it will be cheaper than other things we burn, like coal and gas. This isn't rocket science. Anyone could do it, which is why Europe is looking at it too.
_fizz_buzz_ 6y ago
> electrolysis has a ~45% efficiency That is not correct as far as i know. A modern PEM electrolyzer has an efficiency of around 80%. > All in all, methane would be a saner, simpler target I kind of agree, but this actually the same thing. To synthesis methane, one uses the water and electricity to create hydrogen with an electrolyzer and then add CO2 through hydrogenation to create methane. Advantage of methane is of course the higher density and it's a bigger molecule that is easier to store.
- wazoox 6y ago
  OK, I probably misremembered the different numbers from conversion, compression, transport, etc, but the overall figure was that generation to wheel efficiency was in the realm of 20%, while generation to wheel with batteries is more in the 85% efficiency range.
- timthorn 6y ago
  Yeah, Powerhouse Energy has got a demonstration plant for this running now, taking poor quality plastic waste streams and generating H2 and syngas, as well as heat & electricity: https://www.powerhouseenergy.net/ https://www.powerhouseenergy.net/
- darksaints 6y ago
  There are research SOECs that have exceeded 99% efficiency, owing to the fact that the inefficiency at direct electrical conversion gets expended as heat which lowers the electrical energy levels required for water molecules to split. Basically start with an 80% efficient electrolyzed, but insulate the cell to capture all of the expended heat energy and use it to lower the need for electricity to split the cell. This is possible because of SOEC operating temperature ranges which wouldn't be possible with a PEM cell. I should also add that on the other end of the hydrogen energy cycle, production SOFCs are about 60% efficient at converting hydrogen to electricity, not 45%. And the temperatures of the waste heat are high enough to efficiently convert to electricity, bringing total electrical efficiency up to the 70-80% range. Oh, and they can use hydrocarbon fuels too.
  - 6y ago
jacknews 6y ago
Bloomberg obviously have their own agenda, but I agree, hydrogen seems to be a poor choice. We should be looking at electrification where possible, and otherwise, efficient/environmentally-friendly synthesis of small hydrocarbons that can substitute for fossil fuels, eg butanol. And at the same time researching direct-hydrocarbon fuel cells.
- WJW 6y ago
  There's a bunch of places where you'd still want hydrogen for non-fuel purposes, for example in plastics manufacturing. It's also useful as a reducing agent in steel production. Finally, the higher energy density makes it suitable for larger vehicles like long range trucks. Electrification is preferablem but not everything can be electrified. Also, if you have enough renewable generation capacity to reliably cover peak loads then there are also going to be times when you have massive overproduction. Battery costs for the coming decades will be too high with to buffer multiple weeks of power use so it makes sense to have an "electricity sink" where you can dump any excess. Hydrogen production via electrolysis is an excellent option for that since it's very quick to scale up and down and the end product has many different uses.
  - pfdietz 6y ago
    The big use case is production of ammonia. 6% of all natural gas currently used in the world goes into making hydrogen, largely for ammonia synthesis.
  - nickik 6y ago
    Steel production can be done electrically, no need for hydrogen. Electrical metallurgy is very promising technology. Just replacing what we do now with hydrogen is about as difficult and will end up more expensive. Check out Boston Metal. > Finally, the higher energy density makes it suitable for larger vehicles like long range trucks. Electrification is preferablem but not everything can be electrified. Trucks will be battery electric, its clearly more efficient for 99% of the routes. > Also, if you have enough renewable generation capacity to reliably cover peak loads then there are also going to be times when you have massive overproduction. Renewables will only cover peak loads with the help of massive amounts of batteries in the first place. No amount of solar power will create enough energy to cover the evening peak. > Battery costs for the coming decades will be too high with to buffer multiple weeks of power use so it makes sense to have an "electricity sink" where you can dump any excess. Hydrogen production via electrolysis is an excellent option for that since it's very quick to scale up and down and the end product has many different uses. Battery cost have been coming down fast and the trend is looking to continue. I have huge question about the ideas of trying to store many weeks of energy production, the economics of that are very questionable. And hydrogen is just one of many technologies that can potentially 'solve' that if it even ever is solved. Hydrogen is far from clearly the best way to approach that problem.
agumonkey 6y ago
This vid https://www.youtube.com/watch?v=f7MzFfuNOtY https://www.youtube.com/watch?v=f7MzFfuNOtY kinda says the same I like the H2 idea but it's really more complex than all electric it seems. And also late to the game. ps: to be fair, electric batteries and cars required a lot of faith to evolve and progress... who knows if someone won't come with a few innovations for H2 storage and distribution.
- wazoox 6y ago
  Sure, however basic physic won't change, and H2 will remain the less dense form of fuel (compared to any hydrocarbon). A group of aeronautical engineers calculated that long-haul flight planes running on H2 are impossible for energy density reasons, for instance. So H2 can only work for relatively short flights. Would you set up a whole new parallel infrastructure to cater for short flights only? Probably not.
  - agumonkey 6y ago
    Yeah probably not, I'm just saying the world is complex, sometimes the model and the derived limits are not adequate anymore. Surely lots of people calculated that lithium batteries wouldn't be enough for cars 40 years ago. You see what I mean ?
  - PaulHoule 6y ago
    As it is for rockets, methane is the optimal fuel for aviation https://energynews.us/2013/08/26/midwest/could-natural-gas-fuel-commercial-flights-of-the-future/ https://energynews.us/2013/08/26/midwest/could-natural-gas-f... small airports can get liquid methane the same way they get liquid hydrocarbons https://www.onlocationinc.com/energy-policy-case-studies-liquefied-natural-gas-trucks https://www.onlocationinc.com/energy-policy-case-studies-liq... large airports could have a big pipeline compressor and liquification station. At least liquid methane doesn't take weeks to get into thermodynamic equilibrium as does hydrogen: https://en.wikipedia.org/wiki/Spin_isomers_of_hydrogen https://en.wikipedia.org/wiki/Spin_isomers_of_hydrogen Methane is a better synthesis target than liquid fuels from biological materials, coal, asteroid tar, etc. It would be sweet if you could get the carbon out of the atmosphere somehow.
msandford 6y ago
I believe the hype, but more as hype than actual good ideas. The EU will likely spend a lot of money to at least get started. I would be very surprised if at some point there isn't a change to switch to methane instead. 1. Existing distribution network for natural gas can instead distribute pure synthetic methane as natural gas is 96-99% methane. 2. Vehicles can be adapted to CNG without complete redesign. A 5k retrofit might be attractive vs scrapping a whole car. 3. It would be possible with some work to retrofit airplanes to burn LNG. You might have to lose some of say the baggage area for LNG tanks and insulation but once the engines are running the fact that the chilled LNG is boiling away is A-OK since you want to feed it into your engines anyhow. 4. A lot of existing peaker power plants already burn natural gas, no reason they couldn't burn methane too. The round-trip efficiency might not be great these plants already exist. It might be possible to repurpose them to run perhaps only a few dozen hours per month when the wind and solar in your local region happen to both be low output.
- WJW 6y ago
  A lot of the production methods for methane at scale rely on plentiful available hydrogen though, so this is not necessarily a contradiction.
- gtfoutttt 6y ago
  Yeah the existing infrastructure is the biggie. In the US anyway, we already have a backbone of natural gas pipelines that go to most areas of the country. And CNG is already being used on busses, trash trucks, and tractor trailers every day. I had a CNH civic that was a great little car, and basically identical to the gasoline version except for range. Honda stopped making them due to lack of demand. Bummer.
bjarneh 6y ago
> My unpopular opinion is that H2 is an overall very poor solution to our current problems. That cannot possibly be an unpopular opinion. Hydrogen is way to cumbersome/expensive/dangerous to serve any useful purpose in transitioning to a "Carbon neutral" society.
- manfredo 6y ago
  Hydrogen is probably a feasible power source for container ships. Its energy density by unit of mass is very good, though energy density by volume isn't as good as methane. The cost of hydrogen containment is a function of surface area, and containers ships have large fuel tanks. Cryogenic storage may be feasible for ships, since the ship is surrounded by water which makes heat exchangers work better. Cryogenic storage works by constantly releasing a bit of the fluid to keep temperature down. In the case of a ship crossing the ocean, fuel is constantly being siphoned off to power the boat anyway. Methane is easier to work with, but the sabatier process (synthetic methane) requires a source of carbon. By comparison, it's feasible for costal nuclear power plants to desalinate water and create hydrogen through thermochemical reaction. The energy used in desalination isn't wasted, since bringing the water up to temperature is the first step in thermochemical hydrogen production.
sergeykish 6y ago
> Conventional alkaline electrolysis has an efficiency of about 70%. Accounting for the accepted use of the higher heat value (because inefficiency via heat can be redirected back into the system to create the steam required by the catalyst), average working efficiencies for PEM electrolysis are around 80%. This is expected to increase to between 82–86% before 2030 [1]. It is to store peak renewable energy (negative price), and to achieve Zero Carbon by 2050 (no natural gas). [1] https://en.wikipedia.org/wiki/Electrolysis_of_water#Industrial_output https://en.wikipedia.org/wiki/Electrolysis_of_water#Industri...
- wazoox 6y ago
  Another of my unpopular opinions is that wind power is an abysmally bad capital investment with a terrible ROI (when not heavily subsidized) and even worse EROEI, and all that circus about H2 is a new way to meddle with the financial numbers to paint a rosy (greeny?) future where H2 comes to the rescue before the veil falls. It can only work as long as we have no supply crunch on hydrocarbons. I understand peak renewable at negative prices as an artefact of the market structure. If comes a time when there's actual supply problems on hydrocarbons, we'll need some other sensible way to provide a baseline, and so far only nuclear does. Furthermore, even for H2 production nuclear probably makes more sense than wind power.
  - sergeykish 6y ago
    You said false fact and said nothing about when corrected. Maybe that's the reason your opinion is unpopular.
    - wazoox 6y ago
      I misremembered the actual value, but that doesn't change the overall point that the whole process from power generation to the wheel (in the case of vehicles) have a very poor efficiency, not significantly better than traditional diesel cars and much worse than battery powered vehicles. Plus my opinion appears unpopular in marketing and financial circles, but relatively popular among engineers. Conclude whatever you want. I sincerely believe we'll reach 0% carbon by civilisational collapse, so that's not exactly that I'm against green power, either.
Mvandenbergh 6y ago
Electrolysis efficiency is a lot higher than that although of course the round-trip is much worse than batteries. You can in fact use the existing gas distribution grid although not most of the transmission grid because of material compatibility issues. The one area where I think hydrogen can be hard to beat in areas with the geology for it is as an inter-seasonal energy store in salt caverns. Diurnal storage is obviously a battery play but in a deeply decarbonised system (<10gCO2/MWh) you end up with needing something to provide energy during a few-week-long period of overcast and windless weather (that coincides with the coldest period of the year in NW Europe). For lower levels of decarbonisation, you can just keep your gas generation around and fire them up for a few weeks a year but once you go all the way close to net zero you either need: CCS (unproven, not everywhere has the right geology) Lots of nuclear... but although plants can technically operate in load following mode, the plan economics (low marginal cost, high capital cost) don't support that. Batteries at a scale and a cost that don't seem plausible even with the most aggressive LiOn learning rates H2 storage in salt domes Something else? Part of the problem is that the high capital cost of H2 electrolysis and fuel cells doesn't really support a model where you only produce during the cheapest / windiest 1000 hours of the year and consume during the least windy / most expensive 1000. Really you want an electrolyser fuel combo that has today's so-so performance but low capex, that would be much more compatible with a plan to use it as bulk energy store.
- pfdietz 6y ago
  You go with mediocre efficiency alkaline electrolysers and combustion turbine based generators. The name of the game for this use case is minimizing capital cost.
hannob 6y ago
You're complaining about the low efficiency of H2, and then you end up proposing methane as an alternative? To create methane from electricity you need hydrogen to begin with. To create methane from that you need CO2, which you can get from the air and needs a lot of energy (alternatively you can get it from other emisssion sources, but well, we want to get rid of them, so this is at best an intermediate solution). You end up with much less efficiency compared to hydrogen. Methane has only one thing going for it, and that is existing infrastructure and processes. But that's a mode of thinking where you try to think how you can keep the technology from a fossil energy system. I share some skepticism about hydrogen. It probably shouldn't be applied where more efficient technologies are available. But there's a whole lot of areas where hydrogen really is the only game in town right now. And in many other areas it's an intermediate product needed for further steps.
- NickM 6y ago
  Even if creating methane is less efficient, replacing natural gas with synthetic methane still seems much more practical than trying to replace every single natural gas powered furnace, boiler, hot water heater, dryer, stove, oven, etc. etc. in every home that uses any of these appliances anywhere with electrical equivalents. Hydrogen alone doesn't help solve any of that.
  - yencabulator 6y ago
    Plenty of places in Europe never used natural gas as much as the US does. You're imagining needing to replace infrastructure that doesn't necessarily even exist.

by wazoox 6y ago

usrusr 6y ago

> then compressing the H2 to 700 bars consumes 30% of the stored energy, then you must add distribution and transport (and mostly can't use the huge, existing gas transportation infrastructure). I recall reading somewhere that (in some/many countries) existing gas infrastructure is already specified to tolerate a meaningful (if small) fraction of H2 in the mix and that this fraction isn't currently maxed out, implying that it could be used as a sink for a quite sizable amount of new H2 production capacity as is. Background: historically the network was designed to tolerate a much higher H2 fraction because the initial gas source was a conversion process from coal that happened to yield a H2 fraction considerably higher than what is currently distributed. So as long as conventional gas use continues you'd have zero storage problem for new H2 production until the existing tolerance limit is maxed out. Only switching to anything close to 100% H2 would be a storage/distribution problem. Bonus implication: retooling existing infrastructure back to tolerate that old, higher H2 fraction would be considerably (orders of magnitude) easier than building a network for pure H2. I still share your main view, that there tends to be too much hype for H2. E.g. be wary of people who use an imminent "H2 future" as an argument to abandon all other projects. But that doesn't negate the value of real-life H2 projects at all: chances are we won't satisfy easy H2 use cases any time soon and the things that would make difficult/impossible H2 use cases difficult/impossible shouldn't be used as arguments against production until the easy use cases appear in reach to getting saturated.

wazoox 6y ago
I don't know for other countries, but in France Enedis (in charge of the distribution network) indicated that the network can support no more than 6% of H2. IIRC higher proportions of H2 would actively deteriorate tubes and stuff (I don't know the details, but apparently it can even react with certain materials and make them brittle for instance).
- rjsw 6y ago
  It reacts with fairly common materials [1], steel is the main one. [1] https://en.wikipedia.org/wiki/Hydrogen_embrittlement https://en.wikipedia.org/wiki/Hydrogen_embrittlement
- usrusr 6y ago
  That's roughly the number I was talking about. Subtract current H2 percentage in the system and multiply the result with network throughout (plus storage headroom). That's quite a lot slack to pick up before running out of storage for regenerative H2 production. Sure, those 6% are a local maximum which further limits how much a hypothetical production point could push into the network, but that'd be more of a tactical problem (where to deploy the tech) than a universal limitation. And a solvable one, because the inputs to the conversion are quite transferable.
- hokkos 6y ago
  Enedis is in charge of electricity distribution, GRDF is in charge of fossil gas distribution, and also one of the partners of the grhyd experiment to inject H2 in fossil gas networks, and they went to 20% ! http://grhyd.fr/ http://grhyd.fr/

funcDropShadow 6y ago

And methane has the added benefit of being able to be used in unchanged home heating systems, at least in large parts of Germany and probably of Europe. And there is an existing distribution network in the form of pipelines to most houses. And there are large storage facilities, like the strategic reserves for natural gas. If you factor in the saved CO2 due to the fact that you don't have to change most home heating systems. And that you can easily retrofit lots of gasoline engines in already existing cars. Then I don't get why Germany started to support EV at all.

manfredo 6y ago
The problem is that synthetic methane requires a source of carbon. Extracting the fraction of a percent concentration of carbon in the atmosphere is not feasible. Hydrogen is one of the inputs for the sabatier process, so if the lower energy density by unit of volume isn't a concern it makes more sense to just use the hydrogen directly. The reality, though, is that none of the proposed energy storage solutions are easier than just building nuclear power plants. France already generates most of it's electricity from nuclear power. Belgium recently surpassed 50% nuclear power generation. It's more feasible to just avoid the unsolved energy storage problem altogether.
- epistasis 6y ago
  Some startups are working in direct air capture of CO2. I've linked one in another comment here, Carbon Engineering: https://en.wikipedia.org/wiki/Carbon_Engineering https://en.wikipedia.org/wiki/Carbon_Engineering They think that at scale, a ton of CO2 will cost $100 and take 5-6GJ if energy. Totally agreed that it is far cheaper to just never emit in the first place. But as far as building new nuclear, France has not been able to succeed at this economically, and similarly the UK's plans at building new nuclear are coming apart completely. And even if we are able to decarbonize, with whatever technology, be it nuclear or not, we will still need to get to negative CO2 emissions in the coming decade. While some of that will take the form of CO2 sequestration in the biosphere, such sequestration is only temporary, and we will eventually want geologic sequestration. So we must solve the capture problem by some means. Perhaps it will be with something like Carbon Engineering's process, or maybe something else.

marsokod 6y ago

Regarding the planes, the Soviets had a plane working with hydrogen (and methane I believe). It's not particularly impossible, and at least much more doable than using batteries (for commercial jets). The flying wing is a design improving the efficiency know for a long time (Airbus guys teaching us about it during my studies 15 years ago), but creates additional difficulties, on top of which the impossibility to evacuate the plane fast enough. You would have the same benefits whether your plane is using hydrogen or kerosene. I still do not see planes using hydrogen. Just doing the math on how much electricity would be needed makes it really unlikely.

wazoox 6y ago
In the 70s engineers in CEA in France modified some cars on the campus to run on liquid hydrogen. Apparently burning hydrogen in a combustion engine is pretty straightforward. I think that's the plan about future H2 planes, make them use turbofans slightly modify to burn H2 instead of kerosen. However the big problem is storing H2 onboard. Most of the fuel is in the wings, and very cold liquid hydrogen (icing) as well as very heavy high-pressure tanks won't fit there. That's why I mentioned that an entirely new concept may be necessary to make room for the fuel tanks inside the fuselage.
- rjsw 6y ago
  The Airbus proposals from a couple of weeks ago had the fuel tanks in the rear of the fuselage.
  - wazoox 6y ago
    Yes, that's the problem. You embark much less energy as fuel, lose lots of room for passengers and cargo; I very much doubt this could make sense economically. My brother-in-law who is an aeronautical engineer at Airbus isn't convinced, either.

the8472 6y ago

> then compressing the H2 to 700 bars consumes 30% of the stored energy, then you must add distribution and transport You don't need any of that for grid storage. You can pump it down into a salt cavern in gaseous form (previously used for natural gas storage) and pump it back for on-site power generation when the grid needs it.

pfdietz 6y ago
Also, the energy of compression can be substantially recovered when the hydrogen is used.

pfdietz 6y ago

Your objection misses the mark. There is no singular "energy storage problem". There are various energy storage problems. All hydrogen (or batteries, or other storage technologies) have to do to be viable is address one or more of them. No storage technology can solve all of then, and no single storage technology has to. Hydrogen is best suited for long term and infrequently tapped energy storage. For that use case, it is vastly superior to batteries, even if its efficiency is much lower. And this use case has to be solved to get to a 100% renewable grid. Even aside from that, realize that today 6% of all natural gas consumed in the world goes into making hydrogen, mostly for manufacture of ammonia. So this use ALONE -- which amounts to 700 cubic kilometers of H2 (at STP) per year -- is a vast market for renewable hydrogen production.

jiggawatts 6y ago
You miss his point. Hydrogen is the worst at everything, if compared to all available alternatives, not batteries only. Other technologies outperform it very substantially in almost every metric, often simultaneously. Hydrogen is energy inefficient to produce, space inefficient to store, dangerously explosive, reactive with common engineering metals, and unproven. The one thing hydrogen is optimal for is mass efficiency, but methane is almost as good and far easier and safer to store than hydrogen, and has a much higher energy density by volume, which matters more than you'd think, even for planes and rockets. The new super-heavy rocket designed by Space X uses methane, and the current ones use kerosene. For grid storage, the sole advantage of Hydrogen -- its low density -- is utterly pointless. Nobody cares how heavy building-sized storage is. Efficiency matters, and the electricity-to-hydrogen-to-electricity round trip efficiency is woeful. There's no clear way to make this anywhere near as efficient as batteries either. Not to mention that the huge benefit of batteries is nearly instant (single-digit millisecond!) reaction time, allowing power demand fluctuations to be efficiently smoothed out. Hydrogen can't do that, even with fuel cells. Lastly, we make hydrogen for ammonia to use as fertiliser. This doesn't in any way mean that it this is efficient or worthwhile for power storage.
- pfdietz 6y ago
  > Hydrogen is the worst at everything, if compared to all available alternatives, not batteries only What's better for seasonal storage?

rstuart4133 6y ago

> My unpopular opinion is that H2 is an overall very poor solution to our current problems. I think you opinion is based on dated ideas. The interest in hydrogen is not being driven by electric cars, or as a battery replacement. It's being driven by one thing: if the price of renewable energy keeps dropping, it will be cheaper to make hydrogen than frac gas or mine coal. Even better, one accepted way to deal with renewables intermittency is to over provision. Over provisioning means we produce lots of unwanted electricity that is effectively free. If you can bottle what get get for free and sell it: profit. To put this another way, the hydrogen will be used to produce heat. While lights, cars, planes and other things get all the attention, plain simple heating things is the number one user of energy. We use it to cook food, warm our houses, take showers, wash our plates and pants, iron our clothes, melt silicon ingots, make steel and concrete. Just about everything we use need heat in its production. As you say storing hydrogen is problematic. As possible solution of storing it as anhydrous ammonia has been thwarted by our inability to reverse the reaction and get the hydrogen back. That little sticking point aside, it's nearly ideal. We efficiently produce enormous quantities of it now for fertilizer, and have decades of experience in ensuring your average job farmer handles this rather obnoxious substance safely. But, where there is a lot of money to be made, people have an incentive to go looking. And in Australia at least it dawned there was a _lot_ of money to be made. Australia is the 5th largest exporter of natural gas in the world. Like ammonia natural gas is also difficult to handle. To transport methane efficiently it has to be a liquid, which is done by cooling it to -26O F and compressing 4psi. While exports are great now, it's on the nose. China, Australia's biggest customer, plans to be carbon neutral by 2060. But looming disaster could be circumvented as Australia has huge reserves of sunshine at it is mostly desert, it is politically stable and is very good at things requiring large coordinated infrastructure investment like food production and mining. Hydrogen would be a natural fit for us, if we could solve the storage problem. So, they did. Or more precisely the scientific arm of the Australian government, CSIRO, did. They came us with a metal membrane that converts the ammonia back to hydrogen. https://www.csiro.au/en/Research/EF/Areas/Renewable-and-low-emission-tech/Hydrogen/Hydrogen-membrane https://www.csiro.au/en/Research/EF/Areas/Renewable-and-low-... I guess the plan is instead of taking delivery of a tank of methane, you take delivery of a tank of ammonia, and you convert it back to hydrogen and burn it. Burning hydrogen sounds so low tech, so inefficient. It is low tech, but burning it for heat production is 100% efficient. And remember the driving force behind this is it will be cheaper than other things we burn, like coal and gas. This isn't rocket science. Anyone could do it, which is why Europe is looking at it too.

_fizz_buzz_ 6y ago

> electrolysis has a ~45% efficiency That is not correct as far as i know. A modern PEM electrolyzer has an efficiency of around 80%. > All in all, methane would be a saner, simpler target I kind of agree, but this actually the same thing. To synthesis methane, one uses the water and electricity to create hydrogen with an electrolyzer and then add CO2 through hydrogenation to create methane. Advantage of methane is of course the higher density and it's a bigger molecule that is easier to store.

wazoox 6y ago
OK, I probably misremembered the different numbers from conversion, compression, transport, etc, but the overall figure was that generation to wheel efficiency was in the realm of 20%, while generation to wheel with batteries is more in the 85% efficiency range.
timthorn 6y ago
Yeah, Powerhouse Energy has got a demonstration plant for this running now, taking poor quality plastic waste streams and generating H2 and syngas, as well as heat & electricity: https://www.powerhouseenergy.net/ https://www.powerhouseenergy.net/
darksaints 6y ago
There are research SOECs that have exceeded 99% efficiency, owing to the fact that the inefficiency at direct electrical conversion gets expended as heat which lowers the electrical energy levels required for water molecules to split. Basically start with an 80% efficient electrolyzed, but insulate the cell to capture all of the expended heat energy and use it to lower the need for electricity to split the cell. This is possible because of SOEC operating temperature ranges which wouldn't be possible with a PEM cell. I should also add that on the other end of the hydrogen energy cycle, production SOFCs are about 60% efficient at converting hydrogen to electricity, not 45%. And the temperatures of the waste heat are high enough to efficiently convert to electricity, bringing total electrical efficiency up to the 70-80% range. Oh, and they can use hydrocarbon fuels too.
- 6y ago

jacknews 6y ago

Bloomberg obviously have their own agenda, but I agree, hydrogen seems to be a poor choice. We should be looking at electrification where possible, and otherwise, efficient/environmentally-friendly synthesis of small hydrocarbons that can substitute for fossil fuels, eg butanol. And at the same time researching direct-hydrocarbon fuel cells.

WJW 6y ago
There's a bunch of places where you'd still want hydrogen for non-fuel purposes, for example in plastics manufacturing. It's also useful as a reducing agent in steel production. Finally, the higher energy density makes it suitable for larger vehicles like long range trucks. Electrification is preferablem but not everything can be electrified. Also, if you have enough renewable generation capacity to reliably cover peak loads then there are also going to be times when you have massive overproduction. Battery costs for the coming decades will be too high with to buffer multiple weeks of power use so it makes sense to have an "electricity sink" where you can dump any excess. Hydrogen production via electrolysis is an excellent option for that since it's very quick to scale up and down and the end product has many different uses.
- pfdietz 6y ago
  The big use case is production of ammonia. 6% of all natural gas currently used in the world goes into making hydrogen, largely for ammonia synthesis.
- nickik 6y ago
  Steel production can be done electrically, no need for hydrogen. Electrical metallurgy is very promising technology. Just replacing what we do now with hydrogen is about as difficult and will end up more expensive. Check out Boston Metal. > Finally, the higher energy density makes it suitable for larger vehicles like long range trucks. Electrification is preferablem but not everything can be electrified. Trucks will be battery electric, its clearly more efficient for 99% of the routes. > Also, if you have enough renewable generation capacity to reliably cover peak loads then there are also going to be times when you have massive overproduction. Renewables will only cover peak loads with the help of massive amounts of batteries in the first place. No amount of solar power will create enough energy to cover the evening peak. > Battery costs for the coming decades will be too high with to buffer multiple weeks of power use so it makes sense to have an "electricity sink" where you can dump any excess. Hydrogen production via electrolysis is an excellent option for that since it's very quick to scale up and down and the end product has many different uses. Battery cost have been coming down fast and the trend is looking to continue. I have huge question about the ideas of trying to store many weeks of energy production, the economics of that are very questionable. And hydrogen is just one of many technologies that can potentially 'solve' that if it even ever is solved. Hydrogen is far from clearly the best way to approach that problem.

agumonkey 6y ago

This vid https://www.youtube.com/watch?v=f7MzFfuNOtY https://www.youtube.com/watch?v=f7MzFfuNOtY kinda says the same I like the H2 idea but it's really more complex than all electric it seems. And also late to the game. ps: to be fair, electric batteries and cars required a lot of faith to evolve and progress... who knows if someone won't come with a few innovations for H2 storage and distribution.

wazoox 6y ago
Sure, however basic physic won't change, and H2 will remain the less dense form of fuel (compared to any hydrocarbon). A group of aeronautical engineers calculated that long-haul flight planes running on H2 are impossible for energy density reasons, for instance. So H2 can only work for relatively short flights. Would you set up a whole new parallel infrastructure to cater for short flights only? Probably not.
- agumonkey 6y ago
  Yeah probably not, I'm just saying the world is complex, sometimes the model and the derived limits are not adequate anymore. Surely lots of people calculated that lithium batteries wouldn't be enough for cars 40 years ago. You see what I mean ?
- PaulHoule 6y ago
  As it is for rockets, methane is the optimal fuel for aviation https://energynews.us/2013/08/26/midwest/could-natural-gas-fuel-commercial-flights-of-the-future/ https://energynews.us/2013/08/26/midwest/could-natural-gas-f... small airports can get liquid methane the same way they get liquid hydrocarbons https://www.onlocationinc.com/energy-policy-case-studies-liquefied-natural-gas-trucks https://www.onlocationinc.com/energy-policy-case-studies-liq... large airports could have a big pipeline compressor and liquification station. At least liquid methane doesn't take weeks to get into thermodynamic equilibrium as does hydrogen: https://en.wikipedia.org/wiki/Spin_isomers_of_hydrogen https://en.wikipedia.org/wiki/Spin_isomers_of_hydrogen Methane is a better synthesis target than liquid fuels from biological materials, coal, asteroid tar, etc. It would be sweet if you could get the carbon out of the atmosphere somehow.

msandford 6y ago

I believe the hype, but more as hype than actual good ideas. The EU will likely spend a lot of money to at least get started. I would be very surprised if at some point there isn't a change to switch to methane instead. 1. Existing distribution network for natural gas can instead distribute pure synthetic methane as natural gas is 96-99% methane. 2. Vehicles can be adapted to CNG without complete redesign. A 5k retrofit might be attractive vs scrapping a whole car. 3. It would be possible with some work to retrofit airplanes to burn LNG. You might have to lose some of say the baggage area for LNG tanks and insulation but once the engines are running the fact that the chilled LNG is boiling away is A-OK since you want to feed it into your engines anyhow. 4. A lot of existing peaker power plants already burn natural gas, no reason they couldn't burn methane too. The round-trip efficiency might not be great these plants already exist. It might be possible to repurpose them to run perhaps only a few dozen hours per month when the wind and solar in your local region happen to both be low output.

WJW 6y ago
A lot of the production methods for methane at scale rely on plentiful available hydrogen though, so this is not necessarily a contradiction.
gtfoutttt 6y ago
Yeah the existing infrastructure is the biggie. In the US anyway, we already have a backbone of natural gas pipelines that go to most areas of the country. And CNG is already being used on busses, trash trucks, and tractor trailers every day. I had a CNH civic that was a great little car, and basically identical to the gasoline version except for range. Honda stopped making them due to lack of demand. Bummer.

bjarneh 6y ago

> My unpopular opinion is that H2 is an overall very poor solution to our current problems. That cannot possibly be an unpopular opinion. Hydrogen is way to cumbersome/expensive/dangerous to serve any useful purpose in transitioning to a "Carbon neutral" society.

manfredo 6y ago
Hydrogen is probably a feasible power source for container ships. Its energy density by unit of mass is very good, though energy density by volume isn't as good as methane. The cost of hydrogen containment is a function of surface area, and containers ships have large fuel tanks. Cryogenic storage may be feasible for ships, since the ship is surrounded by water which makes heat exchangers work better. Cryogenic storage works by constantly releasing a bit of the fluid to keep temperature down. In the case of a ship crossing the ocean, fuel is constantly being siphoned off to power the boat anyway. Methane is easier to work with, but the sabatier process (synthetic methane) requires a source of carbon. By comparison, it's feasible for costal nuclear power plants to desalinate water and create hydrogen through thermochemical reaction. The energy used in desalination isn't wasted, since bringing the water up to temperature is the first step in thermochemical hydrogen production.

sergeykish 6y ago

> Conventional alkaline electrolysis has an efficiency of about 70%. Accounting for the accepted use of the higher heat value (because inefficiency via heat can be redirected back into the system to create the steam required by the catalyst), average working efficiencies for PEM electrolysis are around 80%. This is expected to increase to between 82–86% before 2030 [1]. It is to store peak renewable energy (negative price), and to achieve Zero Carbon by 2050 (no natural gas). [1] https://en.wikipedia.org/wiki/Electrolysis_of_water#Industrial_output https://en.wikipedia.org/wiki/Electrolysis_of_water#Industri...

wazoox 6y ago
Another of my unpopular opinions is that wind power is an abysmally bad capital investment with a terrible ROI (when not heavily subsidized) and even worse EROEI, and all that circus about H2 is a new way to meddle with the financial numbers to paint a rosy (greeny?) future where H2 comes to the rescue before the veil falls. It can only work as long as we have no supply crunch on hydrocarbons. I understand peak renewable at negative prices as an artefact of the market structure. If comes a time when there's actual supply problems on hydrocarbons, we'll need some other sensible way to provide a baseline, and so far only nuclear does. Furthermore, even for H2 production nuclear probably makes more sense than wind power.
- sergeykish 6y ago
  You said false fact and said nothing about when corrected. Maybe that's the reason your opinion is unpopular.
  - wazoox 6y ago
    I misremembered the actual value, but that doesn't change the overall point that the whole process from power generation to the wheel (in the case of vehicles) have a very poor efficiency, not significantly better than traditional diesel cars and much worse than battery powered vehicles. Plus my opinion appears unpopular in marketing and financial circles, but relatively popular among engineers. Conclude whatever you want. I sincerely believe we'll reach 0% carbon by civilisational collapse, so that's not exactly that I'm against green power, either.

Mvandenbergh 6y ago

Electrolysis efficiency is a lot higher than that although of course the round-trip is much worse than batteries. You can in fact use the existing gas distribution grid although not most of the transmission grid because of material compatibility issues. The one area where I think hydrogen can be hard to beat in areas with the geology for it is as an inter-seasonal energy store in salt caverns. Diurnal storage is obviously a battery play but in a deeply decarbonised system (<10gCO2/MWh) you end up with needing something to provide energy during a few-week-long period of overcast and windless weather (that coincides with the coldest period of the year in NW Europe). For lower levels of decarbonisation, you can just keep your gas generation around and fire them up for a few weeks a year but once you go all the way close to net zero you either need: CCS (unproven, not everywhere has the right geology) Lots of nuclear... but although plants can technically operate in load following mode, the plan economics (low marginal cost, high capital cost) don't support that. Batteries at a scale and a cost that don't seem plausible even with the most aggressive LiOn learning rates H2 storage in salt domes Something else? Part of the problem is that the high capital cost of H2 electrolysis and fuel cells doesn't really support a model where you only produce during the cheapest / windiest 1000 hours of the year and consume during the least windy / most expensive 1000. Really you want an electrolyser fuel combo that has today's so-so performance but low capex, that would be much more compatible with a plan to use it as bulk energy store.

pfdietz 6y ago
You go with mediocre efficiency alkaline electrolysers and combustion turbine based generators. The name of the game for this use case is minimizing capital cost.

hannob 6y ago

You're complaining about the low efficiency of H2, and then you end up proposing methane as an alternative? To create methane from electricity you need hydrogen to begin with. To create methane from that you need CO2, which you can get from the air and needs a lot of energy (alternatively you can get it from other emisssion sources, but well, we want to get rid of them, so this is at best an intermediate solution). You end up with much less efficiency compared to hydrogen. Methane has only one thing going for it, and that is existing infrastructure and processes. But that's a mode of thinking where you try to think how you can keep the technology from a fossil energy system. I share some skepticism about hydrogen. It probably shouldn't be applied where more efficient technologies are available. But there's a whole lot of areas where hydrogen really is the only game in town right now. And in many other areas it's an intermediate product needed for further steps.

NickM 6y ago
Even if creating methane is less efficient, replacing natural gas with synthetic methane still seems much more practical than trying to replace every single natural gas powered furnace, boiler, hot water heater, dryer, stove, oven, etc. etc. in every home that uses any of these appliances anywhere with electrical equivalents. Hydrogen alone doesn't help solve any of that.
- yencabulator 6y ago
  Plenty of places in Europe never used natural gas as much as the US does. You're imagining needing to replace infrastructure that doesn't necessarily even exist.