[dredmorbius]

Energy storage can be challenging. It is far from impossible.

And of course, all energy conversions involve losses.

Most generally, excess electrical generation can be stored as fuels. The round-trip efficiency is low (~15--20%), but the storage time is proved to multi-hundred-million-year duration.

I've looked into the literature on one variant of this which dates to the early 1960s:

<https://old.reddit.com/r/dredmorbius/comments/28nqoz/electri...>

Your other assertions are ... similarly flawed. Again, yes, challenges, but not outright impossibilities, and there are a number of other alternatives (flow batteries, molten-salt and other batteries, pumped-hydro, CAES) which you fail to consider at all. Several of those are already implemented at grid scale, others ... are at least technically possible, and may well prove viable.

[cryptonector]

Storing a year's worth of U.S. electric production in a way that one could then get it back in a similar time span is not merely challenging. It is infeasible. For one, because of the efficiency issues you mention, we'd need a lot more than a year's worth of electric production to "charge" the "battery", and since we can't dedicate all our energy production capacity to charging that battery, we can't charge it in a reasonable amount of time. Further, we'd need to have the production capacity that could consume that battery, if that battery were fuels made from CO2, say, and we'd have to have it idle except when we need it on an emergency basis -- that or we'd have to use the same kind of fuel as our primary source of electric power under normal conditions, but we're not really allowed to. Moreover, we cannot re-create fissile fuel, so this wouldn't be very dense fuel -- it would have to be chemical fuel meant to be oxidized with atmospheric oxygen. Chemical fuels are not dense enough that we wouldn't notice the storage facilities for them -- they would be enormous, consuming enormous amounts of materials and labor to build. The economic cost of all of that would be staggering -many many times the cost of equivalent amounts of energy that we currently produce-, and that cost is what makes it infeasible.

[dredmorbius]

There's no need to store a year's worth of electrical production. In most cases, a few hours is more than sufficient. Widespread shortfalls are better addressed via grid supply than storage. It might be necessary to have bridging power for a few days, possibly a few weeks, but this is still far from annual-scale in either fraction of production or duration of utilisation.

Synfuels (petroleum analogues), again, though round-trip inefficient are sufficiently energy-dense and long-term storage stable that they might serve for emergency long-term standby capacity. It's useful for other needs (industrial feedstocks and transportation fuels in aviation and marine shipping) so that some supply would likely be necessary regardless. Having standby / idle generation plant would be a capital and maintenance factor, but not impossible. There's well over a century of experience in storing strategic reserves of petroleum in both artificial and naturally-occurring storage facilities.