4 ms·

Nuclear was considerably cheaper when it was built at scale. Both the US during the 70s and France both built plants at a cost of 1-2 billion per GW. Which, in

Nuclear was considerably cheaper when it was built at scale. Both the US during the 70s and France both built plants at a cost of 1-2 billion per GW. Which, in the context of 24/7 carbon-free energy is fantastic.

kragen 6y ago
It was fantastic for the 1970s, but it's not low enough to be competitive with current PV pricing, even with storage.
- sir_bearington 6y ago
  There isn't remotely enough storage available to make renewables at all feasible for decarbonization. To put this in perspective 8 Gigawatt hours is just under one minute of storage for the United States' alone. This is why pointing to the the raw generation costs of intermittent sources is misleading. Generating another 500 MW of solar energy doesn't actually represent any decarbonization if that energy is produced when demand is already saturated.
  - kragen 6y ago
    Apparently you haven't done the math on this, but I did: https://dercuano.github.io/notes/lithium-supplies.html https://dercuano.github.io/notes/lithium-supplies.html Summary of conclusions: It seems likely that known concentrated lithium deposits will not be sufficient to permit the transition to solar over the next decade or two, but there is plenty of lithium in seawater and other, less-concentrated deposits to permit such a transition. New extraction technologies will be needed if lithium batteries are to bridge the intermittency gap. Alternatively, some of the other utility-scale storage technologies might be developed. However, independent of the above, your assertion, "Generating another 500 MW of solar energy doesn't actually represent any decarbonization if that energy is produced when demand is already saturated," is also incorrect. Generating another 500 MW of solar energy to satisfy demand currently being satisfied by coal, oil, or gas allows us to turn off coal, oil, and especially gas plants more often, which does actually represent decarbonization. It's only during hours when 100% of demand is being satisfied by solar energy that generating another 500 MW of solar energy doesn't represent decarbonization — and then only until utility-scale storage or demand response can suck up the zero-marginal-cost energy. Or until carbon emissions are net negative, of course. It's quite plausible for demand response (e.g., charging your Tesla or freezing water in your refrigerator when the sun is shining) to play a bigger part in this than utility-scale energy storage, due to the much lower incremental costs. But my calculations linked above show that there is ample lithium in seawater to do it purely with utility-scale battery storage facilities, even without falling back to pumped storage, demand response, trains full of concrete, compressed-air caverns, etc. Calculemus.
    - epistasis 6y ago
      If we're talking about grid storage, there are many other chemistries with lower specific energy that will not run into resource constraints. For example, sodium may not be a drop in replacement for lithium, but it's far more abundant and makes decent batteries too: https://pubs.acs.org/doi/10.1021/acsenergylett.0c02181 https://pubs.acs.org/doi/10.1021/acsenergylett.0c02181 Then there are entire classes of flow batteries that have not been industrialized yet, that have a lot of potential too. We're using lithium batteries on the grid because we need the high specific energy for EVs, and now that we have an industry it's easy to dual purpose them. But we could also start using those same factories to rolling entire other chemistries that use the same form factor.
      - kragen 6y ago
        That's interesting! I had no idea about sodium batteries! Thank you! Nickel-iron and lead-acid batteries have a lot of historical use for grid-backup applications, and lead-acid ones are still cheaper than lithium, but I don't think the resources are there to scale to an all-solar civilization. Nickel-iron batteries might be cheaper (the resources are a lot more abundant) but nobody sells them anymore I think? There are whole classes of all kinds of energy sources that haven't been industrialized yet, including, to a great extent, nuclear energy. EGS comes to mind. The critical thing about PV is that it's crossed the chasm to mass production and the ensuing price drops. So I think it's crucial that current lithium production won't scale to the needs of an all-solar grid — but that oceanic extraction could. If you liked this thread, you might also enjoy https://news.ycombinator.com/item?id=26219000 https://news.ycombinator.com/item?id=26219000
    - sir_bearington 6y ago
      > Apparently you haven't done the math on this, but I did: https://dercuano.github.io/notes/lithium-supplies.html https://dercuano.github.io/notes/lithium-supplies.html Only 5 minutes of storage exist in known energy reserves. Only 18 minutes exist in lithium reserves accessible with current extraction techniques. How is this supposed to be in favor of renewables? The lesson we take aware here is that demand for lithium needs to drive prices up high enough to incentivize pioneering new mining techniques such that an order-of-magnitude increase in accessible lithium reserves can be tapped. > However, independent of the above, your assertion, "Generating another 500 MW of solar energy doesn't actually represent any decarbonization if that energy is produced when demand is already saturated," is also incorrect. Generating another 500 MW of solar energy to satisfy demand currently being satisfied by coal, oil, or gas allows us to turn off coal, oil, and especially gas plants more often, which does actually represent decarbonization. No, no it doesn't. If you generate 10 GWh of solar during the day and 10 GWh from methane during the night, and you install an additional 10 GWh of daytime solar how much did your carbon emissions go down? Zero. How is this hard to comprehend? > It's only during hours when 100% of demand is being satisfied by solar energy that generating another 500 MW of solar energy doesn't represent decarbonization Which is already the case in some markets. Prophecy that once markets are saturated we'll start using storage to capture the excess is not panning out.

kragen 6y ago

It was fantastic for the 1970s, but it's not low enough to be competitive with current PV pricing, even with storage.

sir_bearington 6y ago
There isn't remotely enough storage available to make renewables at all feasible for decarbonization. To put this in perspective 8 Gigawatt hours is just under one minute of storage for the United States' alone. This is why pointing to the the raw generation costs of intermittent sources is misleading. Generating another 500 MW of solar energy doesn't actually represent any decarbonization if that energy is produced when demand is already saturated.
- kragen 6y ago
  Apparently you haven't done the math on this, but I did: https://dercuano.github.io/notes/lithium-supplies.html https://dercuano.github.io/notes/lithium-supplies.html Summary of conclusions: It seems likely that known concentrated lithium deposits will not be sufficient to permit the transition to solar over the next decade or two, but there is plenty of lithium in seawater and other, less-concentrated deposits to permit such a transition. New extraction technologies will be needed if lithium batteries are to bridge the intermittency gap. Alternatively, some of the other utility-scale storage technologies might be developed. However, independent of the above, your assertion, "Generating another 500 MW of solar energy doesn't actually represent any decarbonization if that energy is produced when demand is already saturated," is also incorrect. Generating another 500 MW of solar energy to satisfy demand currently being satisfied by coal, oil, or gas allows us to turn off coal, oil, and especially gas plants more often, which does actually represent decarbonization. It's only during hours when 100% of demand is being satisfied by solar energy that generating another 500 MW of solar energy doesn't represent decarbonization — and then only until utility-scale storage or demand response can suck up the zero-marginal-cost energy. Or until carbon emissions are net negative, of course. It's quite plausible for demand response (e.g., charging your Tesla or freezing water in your refrigerator when the sun is shining) to play a bigger part in this than utility-scale energy storage, due to the much lower incremental costs. But my calculations linked above show that there is ample lithium in seawater to do it purely with utility-scale battery storage facilities, even without falling back to pumped storage, demand response, trains full of concrete, compressed-air caverns, etc. Calculemus.
  - epistasis 6y ago
    If we're talking about grid storage, there are many other chemistries with lower specific energy that will not run into resource constraints. For example, sodium may not be a drop in replacement for lithium, but it's far more abundant and makes decent batteries too: https://pubs.acs.org/doi/10.1021/acsenergylett.0c02181 https://pubs.acs.org/doi/10.1021/acsenergylett.0c02181 Then there are entire classes of flow batteries that have not been industrialized yet, that have a lot of potential too. We're using lithium batteries on the grid because we need the high specific energy for EVs, and now that we have an industry it's easy to dual purpose them. But we could also start using those same factories to rolling entire other chemistries that use the same form factor.
    - kragen 6y ago
      That's interesting! I had no idea about sodium batteries! Thank you! Nickel-iron and lead-acid batteries have a lot of historical use for grid-backup applications, and lead-acid ones are still cheaper than lithium, but I don't think the resources are there to scale to an all-solar civilization. Nickel-iron batteries might be cheaper (the resources are a lot more abundant) but nobody sells them anymore I think? There are whole classes of all kinds of energy sources that haven't been industrialized yet, including, to a great extent, nuclear energy. EGS comes to mind. The critical thing about PV is that it's crossed the chasm to mass production and the ensuing price drops. So I think it's crucial that current lithium production won't scale to the needs of an all-solar grid — but that oceanic extraction could. If you liked this thread, you might also enjoy https://news.ycombinator.com/item?id=26219000 https://news.ycombinator.com/item?id=26219000
  - sir_bearington 6y ago
    > Apparently you haven't done the math on this, but I did: https://dercuano.github.io/notes/lithium-supplies.html https://dercuano.github.io/notes/lithium-supplies.html Only 5 minutes of storage exist in known energy reserves. Only 18 minutes exist in lithium reserves accessible with current extraction techniques. How is this supposed to be in favor of renewables? The lesson we take aware here is that demand for lithium needs to drive prices up high enough to incentivize pioneering new mining techniques such that an order-of-magnitude increase in accessible lithium reserves can be tapped. > However, independent of the above, your assertion, "Generating another 500 MW of solar energy doesn't actually represent any decarbonization if that energy is produced when demand is already saturated," is also incorrect. Generating another 500 MW of solar energy to satisfy demand currently being satisfied by coal, oil, or gas allows us to turn off coal, oil, and especially gas plants more often, which does actually represent decarbonization. No, no it doesn't. If you generate 10 GWh of solar during the day and 10 GWh from methane during the night, and you install an additional 10 GWh of daytime solar how much did your carbon emissions go down? Zero. How is this hard to comprehend? > It's only during hours when 100% of demand is being satisfied by solar energy that generating another 500 MW of solar energy doesn't represent decarbonization Which is already the case in some markets. Prophecy that once markets are saturated we'll start using storage to capture the excess is not panning out.