3 ms·

> Above certain power requirements, you essentially need the power generation to be onsite. Is this due to transmission losses or just because you couldn't fea

by lurk2 2y ago

> Above certain power requirements, you essentially need the power generation to be onsite. Is this due to transmission losses or just because you couldn't feasibly build enough capacity cables to transmit large amounts of power over long distances?

jandrewrogers 2y ago
A single ton of hydrogen requires ~50 MWh of electricity. Small special-purpose ammonia plants, which are common for some industrial applications, typically require on the order of 50 tons of hydrogen per day. This would require ~2.5 GWh of electricity per day via electrolysis. To put that in context, that is in the same ballpark as the average output of the largest solar farms ever built in the US. The largest ammonia plant in the US requires around 2,000 tons of hydrogen per day. That would require 100 GWh per day to produce by electrolysis, which would require the entire output of a large hydroelectric or nuclear power plant, much like large aluminum refineries. Otherwise, you need to move a lot of electricity or a lot of hydrogen to have good efficiency, and there is infrastructure for neither. Converting natural gas into hydrogen is also energy intensive. One of the big advantages of natural gas is that your hydrogen source is also your energy source and there is vast infrastructure for moving natural gas around. Building green hydrogen pipelines likely makes more sense than trying to backhaul electricity from diffuse sources. A single hydrogen pipeline can reify a lot of electricity production without the concomitant transmission and management infrastructure.
- foobarian 2y ago
  A ton of hydrogen seems to occupy a cube with a 23 meter side. Wonder if a bunch of those could be built to hold the excess gas for night time operation. 100GWh is not small, but it's not impossible. The largest solar farm in operation is 5GW, and that could get you theoretically halfway there operating 10 hours a day. It feels like the challenges are a lot easier to solve than with fusion or nuclear.
  - lurk2 2y ago
    The issue doesn't appear to be storage, but transmission. Hydrogen can leak through metal and lead to it becoming brittle, so you can't use conventional natural gas pipelines to transport it.
    - foobarian 2y ago
      Exactly, these tanks can be giant balloons right next to the solar panels all feeding the in-situ ammonia plant.
      - lurk2 2y ago
        That was my thinking, but I think what he is saying is that power plants won't generate enough electricity to make building an in-situ ammonia plant economical. You need to network power plants together to operate a centralized ammonia plant 24/7, and the network to move this energy (whether in the form of hydrogen or electricity) doesn't currently exist.
        foobarian 2y ago
        The land area of an all-in-one plant is maybe the biggest unknown for me with respect to just getting ownership and permits and such. But it's fun to imagine just picking a giant plot somewhere in the desert and plopping down 20 GW of solar panels, enough hydrogen storage to keep the less energy intensive steps operating throughout the night, and presumably batteries for whatever still requires electric power while panels are offline. Water and air in, sweet sweet ammonia out. :-) Cloudy weather would be an interesting problem I guess.
  - jandrewrogers 2y ago
    It is doable in theory but would require the construction of large-scale supporting infrastructure that currently doesn't exist. I am not optimistic about our ability to undertake infrastructure projects of this magnitude without it taking several decades and incurring obscene cost overruns that make even the most pessimistic economic models look optimistic. This will definitely be harder than nuclear. The expansive land use requirements means the legal battles pertaining to that would almost certainly span many decades. At least with nuclear there is a limited number of people that need to sign-off to have a viable project -- reforming that process probably would be simpler.

by lurk2 2y ago

jandrewrogers 2y ago

A single ton of hydrogen requires ~50 MWh of electricity. Small special-purpose ammonia plants, which are common for some industrial applications, typically require on the order of 50 tons of hydrogen per day. This would require ~2.5 GWh of electricity per day via electrolysis. To put that in context, that is in the same ballpark as the average output of the largest solar farms ever built in the US. The largest ammonia plant in the US requires around 2,000 tons of hydrogen per day. That would require 100 GWh per day to produce by electrolysis, which would require the entire output of a large hydroelectric or nuclear power plant, much like large aluminum refineries. Otherwise, you need to move a lot of electricity or a lot of hydrogen to have good efficiency, and there is infrastructure for neither. Converting natural gas into hydrogen is also energy intensive. One of the big advantages of natural gas is that your hydrogen source is also your energy source and there is vast infrastructure for moving natural gas around. Building green hydrogen pipelines likely makes more sense than trying to backhaul electricity from diffuse sources. A single hydrogen pipeline can reify a lot of electricity production without the concomitant transmission and management infrastructure.

foobarian 2y ago
A ton of hydrogen seems to occupy a cube with a 23 meter side. Wonder if a bunch of those could be built to hold the excess gas for night time operation. 100GWh is not small, but it's not impossible. The largest solar farm in operation is 5GW, and that could get you theoretically halfway there operating 10 hours a day. It feels like the challenges are a lot easier to solve than with fusion or nuclear.
- lurk2 2y ago
  The issue doesn't appear to be storage, but transmission. Hydrogen can leak through metal and lead to it becoming brittle, so you can't use conventional natural gas pipelines to transport it.
  - foobarian 2y ago
    Exactly, these tanks can be giant balloons right next to the solar panels all feeding the in-situ ammonia plant.
    - lurk2 2y ago
      That was my thinking, but I think what he is saying is that power plants won't generate enough electricity to make building an in-situ ammonia plant economical. You need to network power plants together to operate a centralized ammonia plant 24/7, and the network to move this energy (whether in the form of hydrogen or electricity) doesn't currently exist.
      - foobarian 2y ago
        The land area of an all-in-one plant is maybe the biggest unknown for me with respect to just getting ownership and permits and such. But it's fun to imagine just picking a giant plot somewhere in the desert and plopping down 20 GW of solar panels, enough hydrogen storage to keep the less energy intensive steps operating throughout the night, and presumably batteries for whatever still requires electric power while panels are offline. Water and air in, sweet sweet ammonia out. :-) Cloudy weather would be an interesting problem I guess.
- jandrewrogers 2y ago
  It is doable in theory but would require the construction of large-scale supporting infrastructure that currently doesn't exist. I am not optimistic about our ability to undertake infrastructure projects of this magnitude without it taking several decades and incurring obscene cost overruns that make even the most pessimistic economic models look optimistic. This will definitely be harder than nuclear. The expansive land use requirements means the legal battles pertaining to that would almost certainly span many decades. At least with nuclear there is a limited number of people that need to sign-off to have a viable project -- reforming that process probably would be simpler.