4 ms·

Using hydrogen made from electrolysis of water (using solar/wind inputs) as a fuel for electricity generation is basically a futile cycle, and the only advantag

Using hydrogen made from electrolysis of water (using solar/wind inputs) as a fuel for electricity generation is basically a futile cycle, and the only advantage would be temporal storage. For short term storage (hours/days) batteries are far more efficient. For long term storage, hydrogen is a bad idea other than as feedstock for production of longer-chain hydrocarbons (Fischer-Tropsch synthesis) or methane (Sabatier reaction). The more obvious immediate use (not mentioned in article) is for steel production by the H2-DRI-EAF method, the direct reduction of iron in an electric arc furnace. https://bellona.org/news/industrial-pollution/2021-05-hydrogen-in-steel-production-what-is-happening-in-europe-part-two https://bellona.org/news/industrial-pollution/2021-05-hydrog... > "Iron ore is reduced with hydrogen while in a solid state, hence the name direct reduction, to produce direct reduced iron (DRI) called sponge iron. Sponge iron is then fed into an EAF, where electrodes generate a current to melt the sponge iron to produce steel. Some carbon is needed so that steel can be produced. This carbon can come from pulverized coal, biomethane or other biogenic carbon sources." The other obvious use is as a replacement for natural gas feedstocks for ammonia production via the Haber-Bosch process, and there are similar uses in traditional petrochemical production where hydrogen is often used in various large-scale organic chemistry transformations with names like 'hydrocracking' (breaking apart long-chain molecules) and 'dearomatization' (converting benzene to cyclic hydrocarbons), see: https://www.linde-gas.com/en/processes/petrochemical-processing-and-refining/hydrogen_applications_refineries/index.html https://www.linde-gas.com/en/processes/petrochemical-process... Direct air capture of carbon dioxide and conversion to methane via addition of hydrogen (Sabatier process) as a means of making rocket fuel for Starship, that's also an idea Musk should take up. That's how you'd have to fuel a spaceship on Mars as well.

carbonguy 4y ago
> the only advantage would be temporal storage. This is not an insignificant advantage, to be fair. However, I agree that using electrolytic hydrogen as a Fischer-Tropsch or Sabatier feedstock is the way to go.
shawndrost 4y ago
DRI is great! I am [CEO at a hydrogen startup, and] very bullish on hydrogen in the power sector, at least in North America. I should also say, I agree that batteries and "electrify everything" will play a much larger role in the electricity sector vs hydrogen. You've noted that temporal storage is a value prop of hydrogen. I'd like to add on that this is a real value prop driving real $B-scale spend at utilities in eg Los Angeles [1]. Depending on the researcher, hydrogen usually looks like the cheapest form of seasonal energy storage (vs eg iron oxide or flow batteries, excluding pumped hydro in ideal sites). Hydrogen can move electricity around cheaply in space, and is arguably cheaper than HVDC for long distances [2]. If hydrogen can move electricity in time and space, that is kind of the holy grail for wind and solar. As you may know, there are endless square miles of CA/AZ desert and OK panhandle we we can produce power at insanely cheap prices, but we do not, because there is no cheap way to move power around in time and space. (And, because America can't permit those HVDC lines. We're much better at permitting pipelines.) If you believe in consensus cost curves for solar/electrolysis, you probably think that hydrogen will get cost-competitive with natgas @ $8/mmBTU in the next decade or so, then keep dropping. That's basically yesterday in utility years! I'd also list "reliability" as a unique value prop, which is a key goal for grid operators with mandated zero-carbon timelines. This is a one of the things driving Los Angeles utility's acquisitions and cost/risk modeling. (There are other major factors: Utilities own lots of turbines and are building more, and can mostly start using hydrogen today alongside natural gas, and turbines fit their business model and operational expertise.) For these reasons and others, models of zero-carbon power systems are usually more expensive without hydrogen. That is, wind/solar/geothermal/hydro/nuclear/etc, plus 4-12h storage/batteries, tends to be more expensive than the same mix plus hydrogen. Example study at [3]. "24/7 clean energy" is another related concept, and is the holy grail for corporate energy buyers like Google. I haven't even brought up wind/solar interconnection yet, which is another problem which I see driving momentum for hydrogen, but that'll have to wait for another rant. Thanks for reading! [1] https://ww2.arb.ca.gov/sites/default/files/2020-07/ladwp_cn_fuels_infra_july2020.pdf https://ww2.arb.ca.gov/sites/default/files/2020-07/ladwp_cn_... [2] https://www.sciencedirect.com/science/article/pii/S2211467X21000444 https://www.sciencedirect.com/science/article/pii/S2211467X2... [3] https://www.nrel.gov/docs/fy21osti/79444-ES.pdf https://www.nrel.gov/docs/fy21osti/79444-ES.pdf
csee 4y ago
> For short term storage (hours/days) batteries are far more efficient. Is that expected/known to be true even as the tech gets cheaper in 5-10 years? What's the level of confidence of this statement?
Gibbon1 4y ago
> is for steel production by the H2-DRI-EAF method You can electrowin iron at about 80-90% efficiency.

carbonguy 4y ago

> the only advantage would be temporal storage. This is not an insignificant advantage, to be fair. However, I agree that using electrolytic hydrogen as a Fischer-Tropsch or Sabatier feedstock is the way to go.

shawndrost 4y ago

DRI is great! I am [CEO at a hydrogen startup, and] very bullish on hydrogen in the power sector, at least in North America. I should also say, I agree that batteries and "electrify everything" will play a much larger role in the electricity sector vs hydrogen. You've noted that temporal storage is a value prop of hydrogen. I'd like to add on that this is a real value prop driving real $B-scale spend at utilities in eg Los Angeles [1]. Depending on the researcher, hydrogen usually looks like the cheapest form of seasonal energy storage (vs eg iron oxide or flow batteries, excluding pumped hydro in ideal sites). Hydrogen can move electricity around cheaply in space, and is arguably cheaper than HVDC for long distances [2]. If hydrogen can move electricity in time and space, that is kind of the holy grail for wind and solar. As you may know, there are endless square miles of CA/AZ desert and OK panhandle we we can produce power at insanely cheap prices, but we do not, because there is no cheap way to move power around in time and space. (And, because America can't permit those HVDC lines. We're much better at permitting pipelines.) If you believe in consensus cost curves for solar/electrolysis, you probably think that hydrogen will get cost-competitive with natgas @ $8/mmBTU in the next decade or so, then keep dropping. That's basically yesterday in utility years! I'd also list "reliability" as a unique value prop, which is a key goal for grid operators with mandated zero-carbon timelines. This is a one of the things driving Los Angeles utility's acquisitions and cost/risk modeling. (There are other major factors: Utilities own lots of turbines and are building more, and can mostly start using hydrogen today alongside natural gas, and turbines fit their business model and operational expertise.) For these reasons and others, models of zero-carbon power systems are usually more expensive without hydrogen. That is, wind/solar/geothermal/hydro/nuclear/etc, plus 4-12h storage/batteries, tends to be more expensive than the same mix plus hydrogen. Example study at [3]. "24/7 clean energy" is another related concept, and is the holy grail for corporate energy buyers like Google. I haven't even brought up wind/solar interconnection yet, which is another problem which I see driving momentum for hydrogen, but that'll have to wait for another rant. Thanks for reading! [1] https://ww2.arb.ca.gov/sites/default/files/2020-07/ladwp_cn_fuels_infra_july2020.pdf https://ww2.arb.ca.gov/sites/default/files/2020-07/ladwp_cn_... [2] https://www.sciencedirect.com/science/article/pii/S2211467X21000444 https://www.sciencedirect.com/science/article/pii/S2211467X2... [3] https://www.nrel.gov/docs/fy21osti/79444-ES.pdf https://www.nrel.gov/docs/fy21osti/79444-ES.pdf

csee 4y ago

> For short term storage (hours/days) batteries are far more efficient. Is that expected/known to be true even as the tech gets cheaper in 5-10 years? What's the level of confidence of this statement?

Gibbon1 4y ago

> is for steel production by the H2-DRI-EAF method You can electrowin iron at about 80-90% efficiency.