Natural gas turbines can bypass the need to boil water and drive steam turbines, because the actual combustion of the gas creates the rotational energy directly.

Steam turbines are pretty great, though. Not sure how the rest of a coal plant operates, but the turbines themselves shouldn’t be touching any combustion byproduct.

Food is just solar power stored in organic compounds, made from the CO2 in the air.

Other ideas we’ve tossed around are refrigeration and food preservation, but the problem with those is that they need the power when they need the power, and so it’s not exactly a way to sink excess supply.

It can still be a useful sink at small scales. You could make ice at those times of day if you’re eventually going to need that ice later. It takes a lot more energy to chill something (especially water with its high specific heat and latent heat of fusion) that it takes to hold something at temperature in an insulated space. And then go on and use the ice later so that the need to chill something doesn’t have to be synchronized with the exact moment in time you’re drawing energy from the grid to run a refrigeration compressor.

Same with heating. Some smart water heaters can store thermal energy for later, too, and top off their energy usage for some times of day.

I’m not sure if the scale you’re imagining makes these ideas too small to be worth pursuing.

Solar needs active maintenance, including personnel of varying skills. All projects have ongoing costs, especially if they’re gonna sit outside in the weather.

Better to just compare all costs, across the projected lifespan, and compare replacement costs if one source lasts longer than the other.

Doing all that tends to show that building new nuclear isn’t cost competitive. Not big reactors, not small reactors.

nuclear does better for utilities level power than solar.

Define “better.” Personally, I think nuclear is too expensive to be a current solution. Let all the existing nuclear plants continue out their useful lives, and extend them as feasible, but constructing new nuclear plants is probably not worth the cost, even compared to solar + enough grid scale storage to cover multiple nights of demand even when days are cloudy.

Terrapower just got approval to build their $4 billion, 345-MW reactor. That’s $11.6 million per MW.

NuScale canceled their 462 MW project in Utah when it became clear that the total cost was going to exceed $9 billion. That’s $19.5 million per MW.

Solar plants are about $1 million per MW. Grid scale 4-hour batteries are about $750,000 per MW.

And the costs of solar/batteries keep dropping, while nuclear tends to increase in cost over time.

It might be cheaper in some settings.

For certain food styles, I buy bulk spices sometimes because I don’t like to pay for an entire jar I won’t use, knowing that most of it will go stale by the time I’m through the jar. Being able to buy tiny quantities is sometimes way cheaper.

I’m also mismatched in my conditioner and shampoo remaining where I can buy the matching set and let the difference persist, or I can try to buy a single catch-up bottle of whatever I have excess of, to hope that they even out by the time I get to the bottom of a bottle.

Basically, I can imagine where it might be preferable (for both cost and convenience) to buy an arbitrary amount of something rather than buy a fixed factory container of that thing. I know I already do it for certain things.

Grid scale storage is actively being worked on.

Chemical batteries, like rechargeable lithium ion batteries, are a big part of it. Sodium ion batteries and iron air batteries are coming up, as well.

Somewhat related are rechargeable fuel cells and flow batteries, that similarly store chemical energy that can support two-way charge/discharge cycles.

Gravity storage, like pumping water up into a reservoir and then using it to drive turbines on the way down, or elaborate elevator shaft type systems, can store some energy but require lots of land and material, or require very specific geographic features not commonly found.

Kinetic energy storage, turning lots of heavy flywheels and then recapturing that momentum to produce electricity when needed, is also on the grid (and kinda mimics the rotational inertia of the turbines traditionally synced across the grid).

Some other storage technologies include capacitors, pressurized gas containers, and thermal heat storage with molten salt that can be used to make steam to drive turbines on demand.

But all of these solutions are difficult to scale up to the point where they make a significant difference in addressing the mismatch between supply and demand at different times of day. We gotta do all of it, and right now the most cost effective solution is chemical batteries, so that’s been growing at an exponential rate.

It’s like a dumpster filling up, where you have to pay a waste management company to come haul that stuff away, at least if people can’t find a way to take it off your hands for free.

A gravity storage system that stores about 100 MWh and outputs about 25 MW is much, much larger than the 65 battery containers they’d replace. It stores basically 4 hours worth of energy in what appears to be a large steel and concrete structure 150 m tall (the equivalent height as a 30-40 story building) on a 100m x 100m footprint.

If we’re talking about storing a terawatt hour, then we’d be talking about about 10,000 of these gravity storage systems needing to be built. That’s what I mean by existing technology not really meeting the scale requirements of the problem.

Gravity storage systems all basically suffer from this problem. Water-based solutions need to be sited on favorable geography to have large scale (otherwise water itself isn’t dense enough to compete with concrete and stone and sand).

Meanwhile, storing the same 100 MWh of energy in containerized lithium batteries would basically require a 4x6 stack of 40-foot shipping containers that each can store 4MWh.

We can get there on storage, but we’re talking about decades of planning and implementation, across all technologies, before we can even credibly reach storage representing one whole day’s electricity usage. How many man hours of labor does that engineering and planning and building represent? How much steel, energy, and machinery would these projects use up?

Anyone who talks about this stuff without recognizing the scale involved is basically not serious about solving it. It’s an engineering problem that exists independently of money (and it’s also a money problem, but that part will probably pay for itself because of how valuable a solution to this problem would be).

We have the storage technologies, the only thing missing is money.

When discussing large public projects whose scale is larger than anything before seen, the money is mainly an accounting placeholder for the real resources that need to be expended.

Grid scale storage has been expanding at an exponential pace, but the sheer magnitude of the materials and engineering work that needs to be done to make a dent is pretty huge.

Bloomberg projects that total cumulative installed capacity should hit 2 Terawatt hours by 2035, noting that would represent 8x the number for 2025. But when you compare those numbers to just how much electricity is produced or consumed, with 22,000 TWh per year, we’re talking about demand periods measured in minutes, not even hours, much less days.

At scales large enough to make enough of a dent to show up in global energy stats, we need to recognize that even infinite money would run into the real resource constraints of how much capacity we as a species have for pulling minerals out of the ground, processing them into useful materials, and engineering them to be useful energy storage solutions (whether pumped hydro or other gravitational systems, compressed air, flywheels, or whatever battery or fuel cell chemistries can store energy in an efficient way).

We have some technologies, but need things to improve significantly before storage can actually meet the needs for power that meets demand at any given moment in time. In the meantime, matching supply and demand in real time is a true engineering challenge, not just a monetary challenge.